Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats





Volume XXX JANUARY, 1938 Number 1 



Modulated High-Frequency Recording as a Means of Determin- 
ing Conditions for Optimal Processing 


Recording Tests on Some Recent High-Resolution Experimen- 
tal Emulsions J. O. BAKER 18 

Film Perforation and 96-Cycle Frequency Modulation in Sound- 


High-Speed Motion Picture Photography Applied to Design of 

Telephone Apparatus W. HERRIOTT 30 

Vacuum-Tube Engineering for Motion Pictures 


Recent Developments in Gaseous Discharge Lamps 

Projects of the Committee on Standardization of Theater 

Sound Projection Equipment Characteristics 

John K. HILLARD 81 

Recent Developments in Hill and Dale Recorders 

Methods of Blooping . .F. D. WILLIAMS 105 

New Motion Picture Apparatus 

A Combination Picture and Ultraviolet Non-Slip Printer 

O. B. DEPUB 107 

Cine Kodak Model E. . . L. R. MARTIN 112 

An Amplifier for Camera Blimps 


Current Literature 

Spring, 1938, Convention; Washington, D. C., April 25th-28th, 


Society Announcements 125 





Board of Editors 
J. I. CRABTREE. Chairman 



Subscription to non-members, $8.00 per annum ; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West-Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1938, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 


^President: S. K. WOLF, 100 E. 42nd St., New York, N. Y. 

* Past-President: H. G. TASKER, Universal City, Calif. 

^Executive Vice-President: K. F. MORGAN, 6601 Romaine St., Los Angeles, 


** Engineering Vice-President: L. A. JONES, Kodak Park, Rochester, N. Y. 
*Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 
** Financial Vice-President: E. A. WILLIFORD, 30 E. 42nd St., New York, N. Y. 

* Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 90 Gold St., New York, N. Y. 
^Treasurer: L. W. DAVEE, 250 W. 57th St., New York, N. Y. 


*J. O. AALBERG, 157 S. Martel St., Los Angeles, Calif. 

*M. C. BATSEL, Front and Market Sts., Camden, N. J. 

**R. E. FARNHAM, Nela Park, Cleveland, Ohio. 

*G. FRIEDL, JR., 90 Gold St., New York N. Y. 

*A. N. GOLDSMITH, 444 Madison Ave., New York N. Y. 

**H. GRIFFIN, 90 Gold St., New York, N. Y. 

**A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

*S. A. LUKES 6145 Glenwood Ave., Chicago, 111. 

*Term expires December 31, 1938. 

**Term expires December 31, 1939. 





Summary. The quality of variable-width sound records depends to a great extent 
upon image definition. The requirements for a perfect sound-track are complete 
transparency in the clear portion, complete opacity in the dark portions, an extremely 
sharp boundary between the clear and dark portions, and exact duplication of the 
wave traced upon the track by the galvanometer. 

Distortion is introduced by any change in average transmission in recording 
high-frequency waves. At high densities the average transmission is reduced, and at 
very low densities is increased by the presence of the high-frequency waves. The 
average transmission is compared to the transmission through the film for a 50 per 
cent exposed track without signal. 

It is possible to find a density at which there is little, if any, change in average 
transmission, and this density corresponds to most nearly perfect image definition 
and least distortion. On an original or negative recording, with present commercial 
recording stocks, this density is extremely low, of the order of 0.6 to 0.8. For least 
ground-noise, the negative must be recorded at much higher density. A change in 
average transmission of the negative can be tolerated, since by proper choice of print 
density, minimum distortion in the positive track can be attained. 

A modulated high-frequency recording affords an extremely accurate means of 
determining correct negative and print densities for given conditions of laboratory 
processing. An oscillator, designed for several carrier frequencies, is provided with a 
400-cycle modulator for recording. The modulated carrier is recorded for several 
values of lamp current and processed to several negative densities. Prints are then 
processed to various values of density, and the 400-cycle output measured on suitable 
reproducing equipment. The combination of negative and print densities that gives 
least 400-cycle output indicates the condition for best image definition and least 

Care must be exercised in the design and construction of the oscillator to maintain 
the 400-cycle output at a minimum. 

The quality of variable-width sound records will depend upon how 
closely the requirements for perfect wave-form, low ground-noise, and 
freedom from volume distortion are met. Papers have been pub- 
lished in the JOURNAL* from time to time dealing with one or the 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received October 
11, 1937. 

** RCA Manufacturing Company, Camden, N. J. 
t See Glossary. 

4 J. O. BAKER AND D. H. ROBINSON [J. S. M. p. E. 

other of these requirements. Kellogg 1 in 1935 discussed all three in 
a rather comprehensive manner. It is the purpose of this paper to 
consider the problem of wave-form in detail, and to describe a 
method for determining the conditions for optimal processing. 

In 1927, Hardy, 2 treating the general subject theoretically, stated 
"... the quality of sound reproduced by the variable- width type 
of record does not depend upon the conditions of exposure or develop- 
ment of either the negative or positive." This is essentially true of 
the variable-width system, in contrast to the variable-density sys- 



FIG. 1(4). Increase of distortion with volume; 
(B) introduction of new frequency due to wave-train. 

tern, and is quite true for the low frequencies up to those of the order 
of 4000 cps., provided the exposure or development are not too radi- 
cally different from the correct values. The higher the frequencies 
recorded, the greater the necessity for correct exposure and develop- 

Jones and Sandvik 3 in 1930 mentioned certain factors affecting the 
structure of the photographic image, with which this discussion is 
primarily interested: namely, image contraction, image growth, 
and the mutual action of adjacent images. For high-quality repro- 
duction at 4000 cps. and above, these factors become of considerable 


Maurer, 4 in setting values for the sound negative and positive 
densities, based his consideration upon resolving power and contrast. 
While these factors are necessary in variable-width records for fre- 
quency and volume range, the structure of the photographic image 
must also be considered when high frequencies are recorded. Dim- 
mick 5 in 1931 admirably treated the subject of negative and print 
densities for maximum output at the higher frequencies. Imperfec- 
tion of the wave-form introduces not only harmonics and volume 
distortion at high frequencies, but also extraneous sounds, commonly 
known as "raspiness" or "hash." This distortion bears no frequency 
relation to the recorded frequency, but is dependent upon the change 


FIG. 2. Showing how the structure of the image is affected by image 
contraction, image growth, and mutual action of adjacent images. 

of amplitude and the recurrence of the recorded high frequency. 
Fig. 1 illustrates this statement. A shows how the distortion in- 
creases with increase of volume. B illustrates the introduction of a 
new frequency due to the repetition of the normal build-up and decay 
of a high-frequency wave-train. Since speech and music are made 
up of wave-trains of such type, it can be seen that the distortion 
continuously changes in frequency and amplitude. A mathematical 
treatment of wave-form distortion was given by Cook 6 in 1930 and 
by Foster 7 in 1931, and need not be repeated here. This paper will 
be limited to an illustrative discussion and explanation of experimental 
data for showing how the distortion occurs in processing, and how it 
can be minimized. 

Mees 8 in 1935 stated quite concisely the condition for minimum 
distortion: namely. "The point of minimum distortion occurs when 



the lack of sharpness, due to light scattered by the optical system and 
by the photographic emulsion in the recording process, is compensated 
for by a corresponding spreading of the image in the printing process." 
Dimmick suggested the use of a modulated high-frequency record- 
ing for the practical determination of the distortion introduced by 
spreading of the photographic image. E. P. Schultz designed and 
built the first modulated high-frequency oscillator and Dimmick was 
the first to use it to practical advantage, in the early part of 1936. 
Since then, improvements have been made in the design and con- 
struction of the oscillator and much information has been obtained 
on the behavior of image-spread for various emulsions, different types 


FIG. 3. Effect of exposure upon positive. 

of developer, and printers. Without a doubt, this oscillator is the 
most powerful tool found to date, and is useful not only for studying 
the processing conditions of photographic sound records, but also 
for checking other sources of distortion such as found in amplifiers, 
loud speakers, printers, etc. 

The purpose of this paper is to show only its use for photographic 
sound records. A description and explanation of its operation will 
be given later for the benefit of those who wish to take advantage of 
this method. 

Image Definition. In order to understand fully the problem in- 
volved, it will be desirable to consider the image definition and the 
factors affecting it, together with the requirements for a perfect 
sound-track of the variable- width type. A perfect sound-track 
would be one having complete transparency in the clear portions, 

Jan., 1938] 


complete opacity in the exposed or dense portions, an extremely 
sharp boundary between the clear and dense portions, and an exact 
duplication of the wave traced upon the track by the galvanometer. 
The transparency of the clear portion depends upon the inherent 
properties of the photographic material and fog: the opacity of the 
dense portion depends upon the exposure and development; and the 





FIG. 4. Effect of exposure upon average 

sharpness of the boundary upon the characteristics of the photo- 
graphic emulsion and development, assuming the edge in recording 
to be of perfect sharpness. 

The exact duplication of the wave traced by the galvanometer 
depends upon the image definition in both the negative and print. 
As stated previously, the structure of the image is affected by three 
factors, image contraction, image growth or spread, and mutual 
action of adjacent images. All emulsions exhibit these three factors 
depending upon the exposure and development. Fig. 2 helps to 
visualize this a little more clearly. B is the sine wave as traced by 

8 J. O. BAKER AND D. H. ROBINSON [J. s. M. P. E. 

the galvanometer. A is the condition for underexposure, which 
shows image contraction, and C is the condition for overexposure, 
which shows image growth. An exposure can be found where neither 
contraction nor growth of the image occurs. For convenience, we 
shall hereafter refer to the density at which this occurs as the "bal- 
ance-density." At first thought, this would seem to be the proper 
exposure to use. However, with present available emulsions for 
variable-width sound recording, this balance occurs at too low a 







FIG. 504). Effect of exposure upon transmission of modulated high- 
frequency recording. 

density, and, therefore, does not meet the requirement for a perfect 
sound record. Another factor is the present method of making prints 
from a recorded negative. Due to the image characteristics of the 
positive emulsion and printer slippage, the balance-density of the 
print track made from a recorded track of balance-density would be 
extremely low, probably of the order of 0.4 to 0.5, introducing 
considerable noise and having very low output. 

Therefore, of necessity, image-spread must be introduced in the 
recorded track and then balanced out by the image-spread in the 
printed track. 

Jan., 1938] 



Under- and overexposure have the same effect upon the printed 
positive as upon the negative. Fig. 3 illustrates these conditions. 
X and Y would be obtained from Fig. 2 (B). Y, from Fig. 2 (5), 
would give perfect image definition, but would be too low in density 
for best noise-reduction and maximum output. Z would be ob- 
tained from Fig. 2 (A). Y satisfies the requirements and is obtained 
from Fig. 2 (C). 

Average Transmission. Referring to Fig. 4, and translating image 
definition into light transmission, image contraction results in an 
increase of average transmission, where the average transmission is 
taken for unmodulated half-track, and image-growth results in a 
decrease in average transmission Perfect image definition for the con- 

FIG. 5(5). 

Modulated high frequency, 9000 cycles with 
400 cycles. 

dition of balance-density does not change the average transmission. 
Image definition can be determined in a number of ways utilizing this 
fact of its effect upon the average transmission. One method would 
be to record a half-track and a high frequency, and measure the 
change of average transmission on a microdensitometer. This is not 
a practicable method. The method of measuring the d-c. change 
of photocell current when the recording is played on a reproducer re- 
quires either a d-c. amplifier or a sensitive galvanometer. 

The modulated high-frequency recording provides means for mea- 
suring the change of average transmission in terms of alternating cur- 
rent which can be amplified conveniently and measured on suitable 
reproducing equipment. 

Modulated High Frequency. When the image definition is not 
perfect, the change in average transmission of a high-frequency re- 



cording depends upon the amplitude and recurrence of the high fre- 
quency. If, therefore, a high-frequency note is modulated with a 
comparatively low-frequency note the average transmission of the 
high frequency will vary in accordance with the low frequency. 

In Fig. 5 (A) the modulated high-frequency recording is shown for 
the three conditions of under-, correct, and overexposure. For under- 
exposure, the average transmission is increased proportionally to the 
amplitude of the high-frequency note and varies between maximum 
and minimum in accordance with the low frequency. A similar 
phenomenon takes place with overexposure, except in this case the 


75 DB 




FIG. 6. Measuring circuit, modulated oscillator. 

average transmission is decreased. With correct exposure and 
balance-density, the average transmission is unchanged. Fig. 5 (B) 
is a microphotograph of a modulated 9000-cycle track. 

Any change of average transmission is indicated by playing the 
modulated track on a distortion-free reproducer and measuring the 
low-frequency output through a band-pass filter that attenuates all 
frequencies except the modulating frequency. The output will give 
positive readings for all conditions of under- and overexposure, and 
will read a minimum for the condition of correct exposure. Hence, 
the differentiation between over- and underexposure can be deter- 
mined only by curves plotted from a number of readings. 


Graphic Interpretation. -For better understanding and uniform 
interpretation, the following method of plotting the results has been 
adopted. Recordings are made of 1000- and 9000-cycle notes, and 
a 9000-cycle note modulated 75 per cent by a 400-cycle note, of equal 
amplitudes for several values of exposure or recording lamp current, 
with a few inches of unmodulated track at the end of each recording 
for density measurements, and processed in accordance with standard 
practice for variable- width sound negatives. A series of prints is 
then made from each negative and given the standard release print 

FIG. 7, Variable- width sound record image definition characteristics; 
400-cycle output measured through a 400-cycle band-pass filter. 

processing. The prints when measured on a reproducer, using a 
400-cycle band-pass filter for the modulating frequency only, will 
give readings for a number of combinations of negative and print 

The circuits for recording and measuring are shown in Fig. 6. It 
is desirable when recording each frequency, first to adjust the input 
to the recorder for 100 per cent swing of the galvanometer and to 
note the reading of the decibel meter. The input is then reduced by 
1 decibel to avoid the possibility of overshooting. This method of 
setting the input insures uniform amplitude of the recordings. 

In the measuring circuit, the 400-cycle band-pass filter is located 
between the reproducer and the amplifier; otherwise, the output of 

12 J. O. BAKER AND D. H. ROBINSON [J. S. M. P. E. 

the amplifier would saturate the filter and give incorrect readings. 
A switching arrangement is provided for removing the filter from the 
circuit when the 1000- and 9000-cycle notes are being measured. 

In order to provide a uniform method for comparison purposes, the 
output readings must be corrected for all losses appearing in the 
measuring circuit, such as filter attenuation, amplifier response, and 
scanning slit loss. The 400-cycle output must also be referred to 
100 per cent modulation, which, in the case of 75 per cent modulation, 
is the addition of +2.5 decibels. The corrected output readings are 
then all referred to the maximum output at 1000 cycles, as the ref- 
erence level in the decibel system. 

The results are plotted in Fig. 7, with relative levels as ordinates 
and negative densities as abscissas, giving a comparatively smooth 
curve for any one print density. For the condition of optimal proc- 
essing, the 400-cycle output will be a minimum while the 1000- and 
9000-cycle outputs will be at their maxima. 

The 9000-cycle carrier-frequency was chosen in this case merely to 
indicate the method. While preliminary tests indicate that the nega- 
tive and print densities remain the same at any high frequency, it is 
suggested that when the method is used in commercial practice, the 
cut-off frequency of the particular studio equipment under test be 
chosen as the high frequency to be modulated. The lower the cut- 
off frequency, the broader the processing tolerances become. 

The method has been in use for more than a year and a half on both 
the East and West Coasts, and has proved of inestimable value in 
determining the optimal processing conditions for variable-width 
sound recordings. 

In general, the conditions for the optimal processing of ultraviolet 
recordings are a negative density of 1.9-2.1 and a print density of 
1.4-1.6. It must be emphasized that the exact value of negative 
and print density will depend upon the particular processing labora- 
tory and the type and condition of developer used. Experience with 
commercial laboratories has shown that variations of print density 
from 1.4 to 1.8 occur for the same negative density of 1.9. 

The Modulated Oscillator. Since considerable interest has been 
shown in the method and since certain precautions must be observed 
in the construction of a suitable oscillator, it was deemed advisable 
to include here the design of an oscillator that has proved satisfac- 
tory. Unless the oscillator has a very low 400-cycle output from 
the modulated high-frequency source, a satisfactory 400-cycle mini- 

Jan., 1938] 



mum can not be obtained from the recorded sound-track. The 
design given here has a 400-cycle output from the modulated high- 
frequency of - 52 decibels below that of the high-frequency output. 
This level is near that of the ground-noise. 

The modulated oscillator to be described was designed to fulfill 
the following requirements : 



3 400 ^ 

FIG. 8. Modulated carrier oscillator for film measurements. 

(1) Provide high-percentage modulation with minimum distortion. 
(2} Provide means for excluding the 400-cycle modulating component from the 
output circuit. 

(3) Provide several carrier-frequencies. 

(4) Be battery-operated for stability of operation, freedom from hum, and for 

The first and second requirements were fulfilled by utilizing a 
carrier oscillator (6000 to 12,000 cps.), a low-distortion, 400-cycle, 
modulating oscillator, and a balanced modulator. The use of a 


J. O. BAKER AND D. H. ROBINSON [J. s. M. p. E. 

balanced modulator made elaborate filtering unnecessary in the out- 
put and provided linear modulation up to 75 per cent, at which value 
it was used. In order to provide for the several carrier frequencies, a 
switch has been included to alter the capacity across the carrier os- 
cillator tank circuit. 

The complete circuit diagram with its component parts is shown 
in Fig. 8. The carrier oscillator coil, L-l, is of particular importance, 
and has been so designed as to enable an optimal LC ratio to be main- 
tained for the various carrier frequencies from 6000 to 12,000 cps., 
and provides a balanced input to the modulator tube. 

L-2 is the tank inductance for the 400-cycle oscillator, and L-3 is 
the tank inductance for the 1000-cycle oscillator. 










FIG. 9. System for measuring distortion from oscillator. 

A switch, S-l, has been provided for selecting the carrier frequency. 
Switch S-2 is the output selector switch, and is provided with four 
positions: (1) carrier output, (2) modulated carrier output, (3) 400- 
cycle, and (4) 1000-cycle output. The 1000-cycle output is to be used 
as reference frequency. Switch S-3 is the "on-off" switch. Poten- 
tiometer P-l is for varying the output of the frequency selected. 

It will be necessary to use an amplifier having a gain of approxi- 
mately 85 decibels, with a calibrated attenuator reading in decibels 
and providing a flat frequency characteristic to 12,000 cps. The 
output of the oscillator has been adjusted to give approximately 
-50-db. output (0.0125 watt reference level). This output could 
have been increased probably to zero level, but an indicating instru- 
ment would then be required that could read 50, since the range 
required is greater than 50 decibels. A 400-eycle band-pass filter is 
also required, such as the General Radio Co. type 530-A. With 


the value of R-ll given in Fig. 8, a half -scale reading on the voltmeter 
gives an output into a 500-ohm load of approximately 55 decibels 
(0.0125 watt reference level). 

In order to measure the distortion from the oscillator, the following 
procedure is recommended, as shown in Fig. 9. 

(1) Connect the oscillator to the amplifier with a suitable output meter. Set 
the oscillator on carrier (No. 1 position). Set the amplifier attenuator to give an 
output meter reading of approximately 10 decibels. Note the amplifier attenuator 
setting required and the output meter reading for this condition. 

(2) Set the oscillator for modulated carrier (No. 2 position) and connect to the 
400-cycle band-pass filter, the filter output to be connected to the amplifier. 
If the values of R-6, R-7, and R-8 were adjusted properly, no change in oscillator 
potentiometer setting should be required to give the same oscillator meter reading. 
Increase the gain of the amplifier by adjusting the attenuator so that the amplifier 
output meter reads the same as in the first check. The change in attenuator set- 
ting, plus the filter attenuation in decibels, is the attenuation of the 400-cycle 
component to the carrier. This should be down approximately 50 decibels on 
all carrier frequencies. 

Conclusion. Image definition is one of the factors governing the 
quality of variable- width sound records. The control of image 
definition in processing becomes of increasing importance as the 
frequency range is extended. 

Image definition can best be determined in terms of the average 
transmission. Of the several methods for determining average trans- 
mission, the modulated high-frequency recording is the most prac- 
ticable. The density of the exposed track, for best image definition 
and minimum output of the modulating frequency, is referred to as the 
' 'balance-density. ' ' 

The design of a modulated oscillator and a method of using it are 
suggested for general use as an aid in determining the optimal proc- 
essing conditions for variable-width sound records. 

The importance of care in the design and construction of the modu- 
lated oscillator can not be overemphasized. 

Frequent use of the modulated high-frequency recording by a 
number of processing laboratories during the past eighteen months 
has demonstrated the practicability of this method for determining 
conditions for optimal processing. 

The writers wish to express their gratitude to E. W. Kellogg and 
A. C. Blaney for their, helpful suggestions and kindly criticisms in 
the preparation of this paper. 

16 J. O. BAKER AND D. H. ROBINSON [J. s. M. p. E. 


1 KELLOGG, E. W.: "A Comparison of Variable-Density and Variable- Width 
Systems," /. Soc. Mot. Pict. Eng., XXV (Sept., 1935), No. 3, p. 203. 

2 HARDY, A. C.: "The Rendering of Tone Values in the Photographic Re- 
cording of Sound," Trans. Soc. Mot. Pict. Eng., XI (Sept., 1927), No. 31, p. 475. 

3 JONES, L. A. AND SANDIVIK, O.: "Photographic Characteristics of Sound 
Recording Film," /. Soc. Mot. Pict. Eng., XIV (Feb., 1930), No. 2, p. 180. 

4 MAURER, J. A.: "The Photographic Treatment of Variable-Area Sound 
Films," /. Soc. Mot. Pict. Eng., XIV (June, 1930), No. 6, p. 636. 

5 DIMMICK, G. L.: "High-Frequency Response from Variable-Width Records 
as Affected by Exposure and Development," /. Soc. Mot. Pict. Eng., XVII (Nov., 
1931), No. 5, p. 766. 

6 COOK, E. D.: "The Aperture Effect," J. Soc. Mot. Pict. Eng., XIV (June, 
1930), No. 6, p. 650. 

7 FOSTER, D.: "The Effect of Exposure and Development on the Quality of 
Variable- Width Photographic Sound Recording," /. Soc. Mot. Pict. Eng., XVII 
(Nov., 1931), No. 5, p. 749. 

8 MEES, C. E. K.: "Some Photographic Aspects of Sound Recording," /. 
Soc. Mot. Pict. Eng., XXIV (April, 1935), No. 4, p. 285. 


SANDVIK, O.: "A Study of Ground-Noise in the Reproduction of Sound by 
Photographic Methods," Trans. Soc. Mot. Pict. Eng., XII (Sept., 1928), No. 35, 
p. 790. 

STEINBERG, J.: "The Quality of Speech and Music," Trans. Soc. Mot. Pict. 
Eng., XII (Sept., 1928), No. 35, p. 633. 

MILLER, D. C.: "The Physical Properties of Music and Speech," Trans. Soc. 
Mot. Pict. Eng., XII (Sept., 1928), No. 35, p. 647. 

BIELICKE, W. P.: "The Processing of Variable- Width Sound Records in the 
Film Laboratory," J. Soc. Mot. Pict. Eng., XVH (Nov., 1931), No. 5, p. 778. 

SANDVIK, O., HALL, V. C., AND GRIMWOOD, W. K. : "Further Investigations of 
Ground-Noise in Photographic Sound Records," /. Soc. Mot. Pict. Eng., XXII 
(Feb., 1937), No. 2, p. 83. 

DIMMICK, G. L., AND BELAR, H.: "Extension of the Frequency Range of Film 
Recording and Reproduction," /. Soc. Mot. Pict. Eng., XLX (Nov., 1932), No. 5, 
p. 401. 


MR. EDWARDS: How does the scanning slit loss affect the result? 

MR. BAKER: The scanning slit loss reduces the output at the higher frequen- 
cies, the loss depending upon the width of the slit. Of course, the amplifier 
attenuates some of the higher frequencies, too. For a 1-mil slit, say, the slit 
loss at 9000 cps. is 4 decibels compared with the 1000-cycle output. The loss 
increases at the higher frequencies. 

MR. NICHOLSON: The family of curves you showed gave the optimal condi- 
tions, but laboratories are generally interested in limits. How do you pick limits 
from the curves? What determines the point beyond which you may not go? 


MR. BAKER: We like to see the optimal fulfilled, but I am not sure of the limits. 
Present indications are that the modulating frequency should not be greater 
than about 40 db. Our experience in a number of laboratories has been that 
the sharpness of the two sides of one of the family of curves coming down to 
a minimum varies. In some laboratories the spread between the two sides is 
rather large, in other laboratories rather close. 

MR. KELLOGG : Mr. Edwards brought up the question of reproducer slit loss. 
Theoretically a 1-mil slit will give about 65 per cent response, I believe, at 9000 
cps. Of course, in measurements we make allowance for that, which I believe 
is permissible. The calculated slit loss is for a sine wave. 


J. O. BAKER** 

Summary. In another paper 1 it is shown that present commercial sound-recording 
emulsions have least distortion at very low density, accompanied by an undesirable 
amount of ground-noise if used as a positive. 

A new experimental emulsion, E.K. 0-7461-1, differs from present emulsions in 
having extremely high resolution and minimum distortion at a density of approxi- 
mately 1.5. Its speed is less than that of regular recording stocks, but since it is 
used with white light and no filter is required, sufficient densities are readily attained 
with present optical systems. These characteristics offer possibilities heretofore not 

The high resolution, low image spread, and low film-hiss of this emulsion make 
possible recording a positive sound-track that can be played directly, eliminating the 
distortion usually introduced in printing and the ground-noise contributed by the 

The advantage of direct playback will be realized whether the recording is standard, 
Class A push-pull or Class B push-pull. The perfection of image definition in the 
new emulsion means increased processing tolerances in adjusting the Class B system 
for perfect cross-over between the negative and positive half-waves. 

It is not at present feasible to use this emulsion as a negative for making prints on 
Positive stock. For special sound-films without pictures, it may be used for a printed 
positive, using a negative made on the same stock, provided printer losses are not 

It has been shown in an earlier paper 1 how image definition in 
variable-width sound records could be determined by means of modu- 
lated high-frequency recording. For best noise-reduction and 
maximum output, the print density should not be less than 1.3. 
For minimum image distortion this requires a negative exposure that 
will give a negative density of 1.8 to 2.2. 

The photographic problems with which we are confronted in vari- 
able-width sound recording are noise, image definition, and volume 
distortion at high frequencies. 

* Presented at the Fall, 1937, Meeting at New York, N. Y., received Octo- 
ber 11, 1937. 

** RCA Manufacturing Co., Camden, N. J. 




Sandvik 2 shows that the inherent noise of a sound-recording emul- 
sion is quite low, but increases considerably during the process of de- 
veloping, fixing, washing, drying, and handling. Whatever noise 

FIG. 1. Image definition characteristics. 
O ' Experimental emulsion: 13 1 /* min., 65F., developer A. 
X Experimental emulsion: I3 l /z min., 65 F., developer B. 
A Standard emulsion: 10 min., 65 F., developer A. 

White light exposure for experimental emulsion. 

Ultraviolet exposure for standard emulsion. 

exists in the negative is carried over into the print, adding to that 
which results from the imperfections of printing and processing. The 
image definition and high-frequency losses are dependent upon the 



[J. S. M. P. E. 

contrast and resolution of the emulsion. An aid to the reduction of 
these imperfections would be an emulsion that could be reproduced 
directly as a positive without the necessity of making a print, and 
having high contrast and high resolution, with sufficient speed for use 
in present recording systems. The film manufacturers have long 
recognized the need for such an emulsion. Jones and Sandvik 3 in 
1930 described an experimental emulsion with these characteristics, 
















FIG. 2. Sensitometric characteristics; 
experimental emulsion, ISVaimn.; stand- 
ard emulsion, 10 min. Developer A, 
65 F. Positive lamp filter. 

but concluded that further experimental work was necessary before its 
utility and practical value could be determined. 

In the early part of this year samples of an emulsion were supplied 
by the Eastman Kodak Company having a fog value of 0.02, ex- 
tremely free from noise, and having high contrast and high resolution 
together with a high degree of sharpness. The H&D speed of the 
emulsion is relatively low compared with that of present recording 
stocks, but since it was designed for white-light exposure, sufficient 
density can be attained with the present recorder optical systems. Its 
greatest sensitivity lies in the region of 5500 A and hence the only 

Jan., 1938] RECORDING TESTS 21 

change required in the optical system is the removal of the ultraviolet 
filter. This emulsion for the first time offers the possibility of record- 
ing a positive sound-track for direct reproduction. 

In Fig. 1 two sets of image-definition characteristics are shown for 
two slightly different developers, developer A being the one found 
best suited for variable-width recording and developer B having a 
lower concentration of bromide, resulting in a higher speed. The de- 
velopment time was 13 l /z minutes in both cases. Developer A gave 
maximum image definition at a density of 1.45 at a minimum 400- 
cycle level of approximately 50 db. The 9000-cycle level is also a 
maximum, being only 3 db. below that of the 1000-cycle. The den- 
sity range over which the 400-cycle level is more than 40 db. is from 
1.27 to 1.57 or 0.3 

Developer B gave the maximum image definition at a density of 
about 1.05, and the minimum 400-cycle level is again approximately 
50 db., but the density range at 40 db. is only about 0.2. 

The formula for developer A is as follows : 

Elon 0.9 gram 

Sodium sulfite 62 . 8 grams 

Hydroquinone 15.7 grams 

Sodium carbonate (monohydrate) 23 . 5 grams 

Potassium bromide 2 . 1 grams 

Water 1 liter 

The replenisher is the same with the bromide omitted. 
Developer B was mixed the same as A but using only one gram 
per liter of bromide. 

The sensitometric curve of this emulsion developed for 13V2 min- 
utes in developer A is shown in Fig. 2, and the recorder exposure-den- 
sity characteristic in Fig. 3. 

Fig. 4 is a microphotograph comparing the image spread of this 
emulsion with standard recording stock. 

The principal advantage of this emulsion is for direct playback when 
the sound-track is recorded as a positive, and will be realized whether 
the recording is standard, Class A push-pull, or Class B push-pull. 
The inherent perfection of image definition provides increased process- 
ing tolerances in adjusting the Class B system for perfect cross-over 
between the negative and positive half-waves. Recording a positive 
track requires a mask in the recorder optical system that is exactly the 
inverse of the present negative mask. 



[J. S. M. P. E. 

The optical requirements for recording a positive track have been 
worked out for the three types mentioned. The advantage of this 
procedure of using a recorded positive for direct playback or re-re- 
cording is the improvement in sound quality due to the elimination of 
the printing operation, which has been found to introduce a loss of 
image definition even when done under ideal conditions. A substan- 
tial reduction in ground-noise is also achieved, since the printing proc- 
ess always introduces noise at low densities due to pin holes or dirt in 
the negative emulsion and to the graininess along the boundaries of the 
recorded track. 









6-6 7.0 


FIG. 3. Exposure characteristics 

Since the inherent film noise of this emulsion is extremely low, 
ground-noise reduction in the original recording is not essential. 
When used for re-recording to standard emulsions, the ground-noise 
reduction can be applied there. A particular advantage of this 
method would be the ability to anticipate a sudden increase in the 
amplitude of the sound by means of an auxiliary scanning beam ahead 
of the scanning point, which would avoid cutting off the peaks of the 
recorded sound. 

Since the image definition of this emulsion is superior to that of the 
commercial stocks, it is not possible to use it as a negative unless the 
recorded density is made of the order of 2.6 to 2.8, requiring a lamp 
current of approximately 7.2 amperes. Further investigations are 
necessary before the feasibility of this procedure can be determined. 

Jan., 1938] 



In the few tests made using this emulsion for the recorded negative 
and printed positive, it was found that the image definition is materi- 
ally impaired in the printing. More work is required before any defi- 
nite conclusion can be drawn, and the results of the investigation 
will be made the subject of a future paper. 

FIG. 4. Microphotographic comparison of image spread. 

From this discussion, it is seen that the availability of the new emul- 
sion provides a definite advance in the improvement of variable-width 
sound recording, and for the first time permits making a recorded posi- 
tive free of image distortion at usable densities. 


1 BAKER, J. O., AND ROBINSON, D. H.: "Modulated High-Frequency Record- 
ing as a Means of Determining Conditions for Optimal Processing," /. Soc. Mot. 
Pict. Eng., XXX (Jan., 1928), No. 1, p. 3. 

2 SANDVIK, O.: "A Study of Ground-Noise in the Reproduction of Sound by 
Photographic Methods," Trans. Soc. Mot. Pict. Eng., XII (1928), No. 35, p. 790. 

8 JONES, L. A., AND SANDVIK, O. : "Photographic Characteristics of Sound- 
Recording Film," /. Soc. Mot. Pict. Eng., XIV (Feb., 1930), No. 2, p. 180. 


MR. FARNHAM: Was the 9000-cycle record on standard emulsion made with 
ultraviolet light, and the other with white light? 

24 J. O. BAKER 

MR. BAKER: Yes. 

MR. CRABTRBE: It would have been interesting if Mr. Baker could have 
brought along some recordings comparing the method of printing from the nega- 
tive with re-recording from the negative direct. Do I understand that this 
would replace the ultraviolet method of recording? 

MR. BAKER : No, but it is quite useful for special work. We can not print from 
this emulsion to the standard positive emulsion, because the image definition is so 
much better that in order to compensate for the image spread in the negative by 
the image spread in the positive, we have to record the negative at a density that is 
quite high, of the order of 3, to get a print density of 1.4 to 1.5, and we begin to 
lose the higher frequencies at such high negative densities. As we see it now, the 
principal advantage of this emulsion lies in its use in original recording of a positive 
track in the studio, then laying it aside until decision has been made as to what is 
wanted for the final releases; then cutting the negative, running it through the 
film phonograph with the re-recording system, and making the final release nega- 

MR. KELLOGG: How do you account for the disappearance of peaks in the ex- 
perimental emulsion? Why do they not come up as near the top of the swing as 
they do in the standard emulsion? The line of the peaks of the standard emulsion 
is a straight line; while apparently, due to excessive contrast or something, all the 
peaks are ragged in the experimental emulsion. 

MR. BAKER: The peaks grow somewhat in height due to image spread in the 
standard emulsion. In the experimental emulsion, they do not grow quite as 
much. If we could see here on the screen the recorded negatives themselves, I 
think they would look much better than this does. 

MR. KELLOGG: The question has been raised as to whether the optical system 
was equally favorable to both emulsions. Our optical systems are corrected for 
ultraviolet and green. This should be favorable for both films, since in making a 
standard recording we are using exclusively ultraviolet; and in recording on the 
experimental emulsion, the exposure is produced principally by green light. 



Summary. When motion picture film is flexed around a cylinder the film in the 
region of the sprocket holes does not follow a smooth curve. In a sound record this 
leads to frequency distortion of perforation frequency. 

If a length of motion picture film is bent into an arc of short radius, 
it is apparent from visual inspection that the film surface departs from 
a smooth curve in the region of the sprocket holes and assumes a polyg- 
onal profile with the angles of the polygon situated at the sprocket 
holes. This tendency to a polygonal profile of a curved perforated 
film exists even where the arc is of large radius, and is readily apparent 
when suitable methods of observation are used. 

Fig. 1 shows the image of a parallel-line screen reflected from the 
emulsion surface of a length of 35-mm. sound-recording film, curved 
around a cylindrical form of 2-inch radius. Distortion of the line 
images results from distortion of the reflecting surface, the degree of 
distortion indicating the order of departure from the average condi- 
tion. From an examination of Fig. 1 it is evident that distortion of 
the film extends well into the sound-track area, although the picture 
area in general conforms well to the radius. The widening of the line- 
screen pattern shown between sprocket holes in Fig. 1 is evidence of a 
lengthening of the radius of curvature or flattening of the film in this 
area. Distortion of the opposite sign, i. e., shortening of the radius 
of curvature, is shown adjacent to the sides of the perforation that are 
parallel to the direction of film travel. 

That the degree of distortion increases as the radius of curvature 
of form decreases is seen in Fig. 2, where are shown reflected images 
of a series of parallel lines from film surfaces curved over cylindrical 
forms of different radii. It is to be noted that in some cases the dis- 
tortions extend well across the sound-track areas. 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received October 
14, 1937. 

** Bell Telephone Laboratories, New York, N. Y. 



Figs. 1 and 2 were made by means of the arrangement shown in Fig. 
3. For Fig. 1 the axis of the film support was parallel to the optical 
axis of the camera, and for Fig. 2 the axes of the various cylindrical 
forms were at right angles to the camera axis. 

These effects can readily be observed visually by taking a short 
length of motion picture film and viewing the reflected images of the 

FIG. 1. Image of parallel- line screen reflected from emulsion 
surface of film. 

edges or bars of a window from its surface. A trace of oil rubbed over 
the emulsion surface will make it more reflective. 

These figures are evidence that motion picture film curved over a 
sprocket or drum assumes in the region of the perforations a shape 
similar to that shown in Fig. 4. 

If a constant frequency is recorded on a straight length of film, in 
the usual location adjacent to perforations, bending the film over a 
drum or sprocket will change the spacing of the striations of the re- 
cord due to the stretching of the outer, and compression of the inner 



layers. The change of spacing increases as the radius of the drum or 
sprocket decreases. Since differences in curvature exist in the region 
of each perforation, differences in spatial relationship of the sound 
striations (frequency modulation) will exist at perforation frequency. 
Ninety-six cycle frequency "flutter" will therefore be present when 
such a record is reproduced by any means that introduces curvature 
of the film at the point of translation. 

From these observations the following conclusions are reached : 








FIG. 2. Degree of distortion with respect to radius of curvature of film. 

(1) The act of flexing a sound-film record introduces 96-cycle 
flutter if the record is reproduced in the flexed condition. This effect 
is due solely to the film distortion near the perforations resulting from 
flexure, and is not to be confused with effects attributed to the pres- 
ence of the teeth on the drive sprockets. 

(2) No increase in 96-cycle frequency flutter should result from 
the printing operation when conducted in a contact printer, such as 
the Bell & Howell, so long as the pitches of negative and positive are 
such that no slippage occurs. This follows because, in this operation, 
negative and positive are curved together over the same sprocket and 



[J. S. M. P. E. 

so are similarly distorted. Image transfer will therefore be effected 
correctly as far as spatial relationship of the striations is concerned. 
Frayne and Pagliarulo 1 recently reported that the printing operation 
leads to negligible increase in 96-cycle frequency modulation. 




FIG. 3. 

Arrangement for photographing reflections of parallel-line 
screen on film. 

Where pitch mismatching occurs in the printing operation the con- 
tours of negative and positive will not match and an increase in fre- 
quency modulation would be expected, probably at some other than 

FIG. 4. Shape of film curved over a drum, at region of perforations. 

perforation frequency, depending upon the manner of slippage and 
the degree of mismatching. 

(3) Moving the sound-track away from the sprocket holes should 
largely remove this source of frequency modulation. 1 



1 FRAYNE, J. G., AND PAGLIARULO, V.: "The Influence of Sprocket Holes 
upon the Development of Adjacent Sound-Track Areas," /. Soc. Mot. Pict. Eng , 
XXVIII (March, 1937), No. 3, p. 235. 


MR. PALMER: Fig. 2 seems to show a lot of distortion in the 16-mm. film on the 
side opposite the perforations. How would you account for that? 

MR. CRABTREE : We have observed distortions at the side of the film opposite 
the perforations but we have not considered them to be related to 96-cycle fre- 
quency modulation with which this paper is concerned. 

MR. FISHER: What is the quantitative relation between the variation in the 
lines on your figures and the actual amount of variation from a true surface? 
This is an extremely sensitive indicating system, I believe. 

MR. CRABTREE: We have not determined the relation between movements of 
the line pattern and the departures of the film surface from a true plane. 

MR. SACHTLEBEN: Is not such distortion evident even in a flat piece of film 
that has not been distorted by the sprocket holes? 

MR. CRABTREE: General distortions of the film surface are readily apparent 
but are not to be confused with localized distortion around perforations. 

MR. ROBERTS : You stated that printing did not introduce any of this 96-cycle 
flutter. Is it not conceivable, in non-slip printing, that this flutter will occur 
because often the two films are fed in at unequal angles and the point at which 
the printing is done is unsupported? At that point we are engaging, or rather 
contacting, a series of flat surfaces, so in that type of printing why should not the 
effect be serious? 

MR. CRABTREE : That would seem reasonable. 

MR. CARVER: Is it not true that these reflections from surfaces of small curva- 
ture, or large radii of curvature, exaggerate the spaces between the lines very 
greatly, and that the whole effect is one that you can see at any time in film that 
has dried out a little? When the emulsion is shrunken, areas between the per- 
forations are always curved. If the film is dried out to, say, 40 or 50 per cent rela- 
tive humidity, the curvature would be even much greater than you showed. I 
think the effect is very much exaggerated. 

MR. CRABTREE: We are not considering effects due to shrinkage. The 
appearance of shrunken film around perforations is well known. We are con- 
cerned with fresh film used in recording, and the effect upon frequency of flexing 
the film around a cylinder. To illustrate graphically the nature of these local 
distortions, photographs were made using a parallel-line screen. 

MR. CARVER: You do not need the lines to see the distortion. You can see it 
by merely looking at the film. 

MR. CRABTREE : The local distortions are perhaps best revealed by the use of 
straight parallel lines. 

MR. CARVER: Those are reflections of lines that are very much exaggerated, 
and have nothing to do with frequency. 

MR. CRABTREE: The use of a line-screen does not "exaggerate" the nature of 
these distortions. It gives an undistorted although enlarged view of these local- 
ized effects. It was the purpose of this paper to associate these effects with 96- 
cycle frequency modulation. 



Summary. High-speed motion pictures are employed at Bell Telephone Labora- 
tories as a visual aid in the study of problems associated with the design, manufacture, 
and testing of telephone apparatus. A new high-speed camera of the optical compen- 
sator type operating at 4000 pictures per second is described, and its application 
to the study of problems associated with telephone apparatus is discussed. 

High-speed motion picture photography offers to the engineer 
means of analyzing mechanical movements otherwise too rapid to be 
normally perceived. It involves taking a series of pictures on motion 
picture film at high speed and projecting them at a much slower rate, 
such as at the normal viewing rate for motion pictures of sixteen or 
twenty-four pictures per second. In this manner a time-delay factor 
is introduced which is the ratio of the taking frequency to the viewing 
frequency. Pictures taken at the rate of 4000 per second will, when 
projected and viewed at the rate of 20 pictures per second, offer a 
time-delay factor of 200, and a rapid mechanical movement may thus 
appear to take place during a substantially longer period of time and 
so permit a detailed study of the motions involved. 

Photographic records of this type have particular value in that 
space -time relationships of the movements of parts involved may be 
directly determined by measurements either of projected images of 
individual pictures or of enlarged paper prints. The time-interval 
between successive pictures is determined either directly from a clock- 
face photographed at the side of each picture area or from a knowledge 
of the taking rate. 

Another method of analyzing rapid mechanical movement involves 
photographically scanning a small section of the silhouetted image of 
closely adjacent parts of a mechanism that are in relative motion. 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received October 
10, 1937. 

** Bell Telephone Laboratories, New York, N. Y. 



This results in a shadowgraph type of record on sensitized paper or 
motion picture film, from which the space-time relationships can be 
determined. This method, however, imposes severe limitations upon 
the size and the useful area of mechanisms that can be studied, and 
producing silhouetted images of important parts is frequently impos- 
sible due to structural conditions. 

FIG. 1. High-speed camera, operating at 4000 pictures a second. 

High-speed photography is finding extensive application in auto- 
motive and aeronautical engineering, in the study of problems asso- 
ciated with combustion of fuels and vibration in motors, air-flow 
around structures and in propeller design and performance. It is 
used extensively in the study of ballistics and in timing athletic events. 
It is coming into use in the fields of biology and medicine, in the study 
of nervous and muscular reactions under controlled conditions. Its 
widest use is in industry, where it is applied to "motion analysis" of a 
wide variety of manufacturing operations and of problems associated 
with the design and performance of machinery. 

32 W. HERRIOTT [j. s. M. p. E. 

Bell Telephone Laboratories has for several years made use of high- 
speed photography as a visual aid in studying problems associated 
with the design, manufacture, and performance of telephone appa- 
ratus. It has been applied to studies related to dials, relays, switches, 
ringers, and to many other similar devices. It has* found use in a 
study of stress, impact, and noise conditions associated with a variety 
of mechanical or electromechanical units. It has been of particular 
value in the study of transient movement of mechanical elements in 
telephone apparatus. 

Apparatus used at Bell Telephone Laboratories in taking high- 
speed motion pictures includes a specially designed camera, shown in 

Hold Down 


FIG. 2. Schematic arrangement of high-speed camera. 

Fig. 1, which is usually operated at the taking speed of 4000 pictures 
per second. 

The camera is of the optical compensator type having a small cube 
of optical glass rotating at high speed between the camera lens and the 
film, and serves to render the image of the object being photographed 
stationary relative to the rapidly and continuously moving film. 

The compensator cube has four polished faces, each parallel to its 
axis of rotation which is, in turn, parallel to the axis of the sprocket. 
The film passes from the supply spool under a hold-down roller, shown 
in Fig. 2, and is held in contact over half the circumference of the 
sprocket from which it passes to the take-up reel. One picture is 
taken for each quarter revolution of the compensator, and, if blurring 
is to be avoided, the film and the image must move at the same rate. 
The index of refraction of the glass and the dimensions of the cube are 
chosen to cause the correct movement of the image as the film is con- 
tinuously advanced past the exposure area. The downward image 
movement results from rotation of the cube and the consequent 


change in refraction of the light-rays at opposite faces of the cube. 
Rays from a point in the object to the left of the lens are converged by 
the lens and become incident upon the front face of the cube in the 
position at A, at which they are refracted to the opposite face and 
thence pass to form an image of the object point upon the film at a. 
When the film has reached a point b on the lens axis, the compensator 
has rotated to the position B. Displacement of the image is zero at 
this point, but further rotation of the prism causes further downward 
displacement of the image toward the point C, thus allowing the ex- 
posure of an elemental area of the film to continue during a substan- 
tial part of the exposure cycle. The duration of each exposure is con- 

FIG. 3. Interior of the camera. 

trolled both by the speed of rotation of the compensator and by the 
angular height of an aperture in front of each of the four faces of the 

The cube rotates on a ball-bearinged shaft at 60,000 rpm. for a tak- 
ing speed of 4000 pictures per second, and is driven by spur-gears 
from a main shaft directly connected to a V 5 -hp. motor. A toothed 
sprocket drives a 100-ft. length of 16-mm. supersensitive panchro- 
matic film having twice the usual number of perforations, for better 
distribution of stresses in the film during acceleration. The sprocket 
is directly attached to the main drive-shaft of the motor and rotates 
at 12,000 rpm. for a taking speed of 4000 pictures per second. The 
loaded film spool is placed upon the upper spindle as shown in Fig. 3, 
and the film is threaded under a guide roller onto the main sprocket 

34 W. HERRIOTT [J. S. M. P. E. 

and to a take-up spool driven by a separate motor. A finder is pro- 
vided which permits viewing the image on the film as seen upon a 
hooded ground-glass screen mounted upon the hinged door of the 
camera. Lenses of various focal lengths are interchangeable upon the 
front of the camera. The camera is mounted upon a substantial tri- 
pod and is readily portable. 

FIG. 4. Portable lighting units used for high-speed work. 

The effective duration of exposure for each picture is of the order of second or less. It is obvious that intense light-sources must 
be used to illuminate the subject adequately. Portable lighting units 
of the types shown in Fig. 4 have been developed at Bell Telephone 
Laboratories. These employ both carbon arcs and tungsten lamps, 
and offer intensities of illumination of the order of 10,000 to 500,000 
foot-candles as desired. Liquid filters are used for absorbing excess 
heat radiated by the light-sources. 



Facilities for high-speed photography of the type described are used 
as an aid to numerous kinds of development work in the Laboratories 




FIG. 5. Frames from films taken at high speed. 

and are made generally available for this purpose. About 2000 films 
have been made to date, which indicates the extent to which the de- 
vice is finding application to telephone apparatus problems. The 

36 W. HERRIOTT [J. s. M. P. E. 

high degree of portability that has been achieved in both the camera 
and lighting equipment lends itself well to extensive application to a 
wide variety of problems. 

In Fig. 5 are shown series of selected frames from films that are rep- 
resentative of the application of high-speed photography to telephone 
apparatus problems. 

At A is shown the action of the 35A alarm fuse when burn-out 

At B is a study of condenser-can extrusion in a punch-press. Valu- 
able information relating to punch and die adjustment and to the 
flow of metal during the punching operation was derived from a series 
of such pictures. 

At C is shown the burn-out of an experimental exciter lamp from 
which information relating to the influence of variation in coil pitch 
upon lamp life was obtained. 

At D is a series of pictures illustrating the action of the impulse 
wheel, pawl, and snubbing spring in the Type 5 dial. 

At E is shown the action of the clapper striking one gong of an ex- 
perimental 20-cycle ringer. A peculiar acoustical effect was explained 
by this picture which revealed more strokes of the clapper per cycle 
than were desired. 


MR. TOWNSLEY: How do you get your camera up to speed to avoid wasting 
too much film ? 

MR. HERRIOTT: We rely upon the power of the motor to accelerate the 
sprocket rapidly. No speed-regulating or clutch devices are used. We use a 
high-torque motor. 

MR. CRABTREE: Have you done any work with 35-mm. cameras? 

MR. HERRIOTT: Our developments to date have been concerned only with 
16-mm. cameras. 

MR. HARRIS: What lens aperture are you using? 

MR. HERRIOTT: In general, most of our exposures are made at //5 to //8. 
We have gone as low as//20 on certain subjects. Seldom do we go above //4 or 

MR. CRABTREE : Why is it necessary to use two motors? 

MR. HERRIOTT: Our experience has indicated that power applied to the 
sprocket, gearing, compensator, and the supply reel through the film from the 
take-up reel resulted in film breakage and uneven film movement. 

MR. CRABTREE: Could you not connect the take-up by a friction drive? 

MR. HERRIOTT : Our experience has been that a friction drive leads to irregular 
film movement. The take-up motor runs almost without load, and serves only to 
take up slack in the film that has been accurately paid out by the sprocket and its 
drive motor. We experienced considerable trouble in film breakage at the take-up 


reel after the 100 feet of film had passed through the camera. This amounted 
to as much as several yards of film, but has been reduced to a matter of an inch 
or less by the use of a suitable guard around the take-up reel. 

MR. WALKER: How many feet of film are wasted before the film comes up to 

MR. HERRIOTT: This camera will get up to 90 per cent of its speed in about 
30 to 35 feet of film. Then there is a rather slow rise to maximum at the end. 
We are operating at 4000 frames per second and use 100-ft. rolls of film. 

MR. CRABTREE: Are the perforations damaged at all? 


MR. CRABTREE: Perhaps humidifying the film would prevent it from break- 
ing, and permit higher speeds. 

MR. HERRIOTT: That is doubtless true. We like to hold the film rather 
close to the pitch of the sprocket, and find that it will ride off the teeth and trouble 
will develop if the pitch is not reasonably close to that of the sprocket. 

MR. RICHARDSON: What is now regarded as the highest maximum taking 
speed ? 

MR . HERRIOTT : Pictures have been made at 60,000 frames a second by spark 
motion picture photography, which should not be confused with this kind of high- 
speed photography. Special cameras have been built wherein short lengths of 
film can be whirled on a drum. We can not study transient phenomena with 
such a camera, and that is what interests us most. It would be extremely diffi- 
cult to use cameras of such high speed because of the difficulty of synchronizing 
exposure with transient phenomena. Such cameras have, however, found a 
great deal of use in aeronautical studies where steady-state conditions apply. 

MR. CRABTREE: What is the practical limit with your type of camera? 

MR. HERRIOTT: In present practice we are operating at 4000 frames per 
second. We are hopeful of going higher. Perhaps 6000 is the next step. The 
strength of materials employed in the construction of the camera and the strength 
of the film certainly establish an upper limit. Just where that limit lies we can 
not say. We are interested in making pictures at much higher speeds than these 
as we have many applications for the use of higher speeds if they can be proved 
fairly practicable in routine use. 

MR. CR,ABTREE: Does the limit lie in the strength of the film or the mecha- 
nism or both? 

MR. HERRIOTT: We have had little or no trouble with our mechanism at 
4000 frames per second. We have, however, had trouble with film breakage and 
film shrinkage. 



Summary. Manufacturing and developmental technics of vacuum tubes are de- 
scribed with particular reference to their use in motion picture equipment. A brief 
discussion of how application requirements affect the choice of materials, structural- 
design, and electrical characteristics of phototubes and amplifiers of both power and 
voltage types is included. How tubes are designed to meet specific needs is illus- 
trated by reference to recent tube developments. Work on producing tubes having 
low-hum, low -micro phonic, and low-noise characteristics is described as of special 
interest to the motion picture engineer. The paper closes with recommendations as 
to how to use vacuum tubes to best advantage. 

The problems of vacuum tube engineering as applied to motion 
pictures are primarily those of recognizing the special requirements 
that must be solved for the recording and reproduction of sound 
motion pictures. In a few words, these requirements consist of the 
faithful reproduction in the theater of sounds created before the 
microphone when the picture was originally recorded. They are 
not unlike the requirements encountered in the design of public ad- 
dress systems and the audio amplifiers of modern radio sets. The 
main differences are of degree rather than as representing basically 
new problems. The problems of the phototube designer will be dealt 
with in another section of this paper. 

Let us now look at some of the problems facing the tube design 
engineer. One of the major problems is that of reducing the random 
noise that appears in the grid circuit of the first amplifier tube. Since 
this noise is due to discrete changes of electrical potential of the grid 
of the tube and since the function of the tube is to amplify 
such changes of potential, if becomes the problem of the design engi- 
neer to keep these changes as low as possible in order to realize the 
maximum of gain from any given tube structure. There are several 
sources of this noise and it is with each of these that the tube engineer 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received October 
12, 1937. 

** Radiotron Division, RCA Manufacturing Co., Inc., Harrison, N. J. 



is concerned. First is thermal agitation, or the noise resulting from 
the flow of current in the elements of the input circuit of the tube as 
well as within the grid material. Thermal agitation is a function of 
the conductivities of the various metals employed in the tube and the 
tube circuits. For example, since the grid side-rods conduct heat 
away from the grid wires, the materials used for the grid-support 
side-rods must have good conductivity. Also, the engineer must 
balance improved performance against increased difficulties in manu- 
facturing technic. He can not choose a grid wire material that will 
be so soft or so hard that it can not be used in manufacture of the 
finished product because, if he does, the cost of manufacture may be 
out of all proportion to the improvement from the noise-generation 
standpoint. In other words, some compromise in performance must 
be made in order to produce a tube that will economically justify it- 
self. The amount of thermal-agitation noise depends also upon the 
width of the frequency band passed by the amplifier. With the ad- 
vent of high-fidelity recording and reproduction, new lower limits of 
noise must be met. 

A second source of noise within the tube results from the fact that 
the stream of electrons flowing from the cathode to the plate is made 
up of a series of particles rather than a continuous fluid. The electron 
flow to the plate is somewhat irregular, resembling the fall of hail- 
stones. This irregularity gives rise to irregularities in the plate cur- 
rent of the vacuum tube, and hence causes noise in the amplifier. 
Fortunately, it happens that the more electrons available from the 
cathode, the less irregular is the electron flow to the plate. It is there- 
fore very important that the electron emission from the cathode be 
ample if the noise is to be kept low. As a result, the need for low- 
noise tubes has added another reason for improved cathode emission. 

In addition to the causes of noise mentioned above are others such 
as noise arising in the plate curcuit as a result of irregularities in sec- 
ondary emission from the plate, and noise produced by disturbances 
of the space-charge around the cathode due to ionization. The first 
of these can be controlled by choice of materials used in the plate 
structure and also by coating the plates with materials (such as car- 
bon) that have low secondary-emission properties. The second can 
be reduced in manufacture by making the residual gas content of the 
tube exceedingly low by using a very active getter and by accurately 
controlling the heating of the tube electrodes during exhaustion. 
Both these factors are under the control of the design engineer, but 

40 L. C. HOLLANDS AND A. M. GLOVER [j. s. M. P. E. 

he must exercise his judgment and experience to attain the desired 
results by the least costly methods. 

A fourth source of noise is the small leakage that takes place across 
the parts used to support the various electrodes of the tube structure 
and for spacing the electrodes so as to produce the desired electrical 
characteristics. These parts of the tube are the glass supports for 
the lead wires and the mica supports for the grid side-rods, the cath- 
ode, the plate side-rods, and the screen side-rods. Each of these 
parts must be treated in a different manner. For instance, it has 
been determined that some kinds of glass are not as good insulators 
as others. Therefore, glass of only certain compositions should be 
used for supporting the various tube electrodes. On the other hand, 
the glass must be capable of being handled in the manufacturing proc- 
esses with a certain ease and rapidity in order that production will 
not be overly difficult or slow. In the case of mica, it has been found 
that supplies from different parts of the world vary greatly in insulat- 
ing qualities. Also, these qualities may change due to the tempera- 
tures to which they are subjected during the exhaust process. Again, 
the mica must not give off excess moisture during the exhaustion 
process as this tends to affect both the emission from the cathode and 
the amount of residual gas, and hence the extent of ionization that 
may occur in the tube under operating conditions. 

Another problem confronting the tube design engineer is reduction 
of microphonics. Microphonics are the result of physical movement 
of one or more of the electrodes within the tube structure. As a re- 
sult, the designer is faced with the problem of making the electrodes 
of the tube as non-resonant as possible. The most immediate solu- 
tion of this problem would seem to be to make the tube electrodes as 
large and heavy as possible. This, however, is not always practicable 
because we must at the same time obtain the desired electrical char- 
acteristics. This requirement may limit the size of the control grid 
wire, the diameter of the control grid, and the size and shape of the 
plate. Furthermore, it is necessary to provide clearance in insulators 
in order to compensate for expansion as the tube parts become warm 
in service. If tube electrodes are supported too rigidly they may 
buckle or bow out of line and cause changes in the electrical character- 
istics. The buckling may actually go so far as to cause a direct short- 
circuit under operating conditions. It has been demonstrated in the 
laboratory that looseness of the heater in the cathode causes micro- 
phonic noise. As a result of this discovery, it has been found neces- 

Jan., 1938] 



sary to make the heater a force-fit in the cathode, which is a difficult 
problem from the manufacturing standpoint. 

Still another problem that confronts the design engineer is that of 
reducing hum produced by the alternating current used to heat the 
filament of the tube. The hum may be produced by electromagnetic 
effects, or by direct emission from the heater to some element within 
the tube structure, or by leakage between the heater and the cathode. 
Taking the causes one at a time and analyzing them, we can deter- 

FIG. 1. Type 1603 tube. 

mine the method of solution as applied in modern practice. In the 
case of electromagnetic effects, the double-helix method of winding 
the heater is very effective. One leg of the heater is wound in the 
form of a helix for the proper length for a given size of cathode, and 
then the wire is doubled back upon itself to form a bifilar helical type 
of winding. It can readily be seen that in this type of winding any 
electromagnetic field produced in one leg of the winding will be can- 
celled by the electromagnetic field of the adjacent wire. After the 
ends of the heater leave the cathode to go to the support wires in the 
stem they separate. As a result of this separation, the fields sur- 

42 L. C. HOLLANDS AND A. M. GLOVER [j. s. M. p. E. 

rounding these wires no longer cancel. Now if, for example, a con- 
trol-grid support or lead-wire is placed in the stem press close to this 
heater lead, there is always the possibility of introducing alternating 
current into the grid of the tube as a result of the latter's proximity 
to the heater leads. Again, it is advisable to isolate the plate and 
other electrode leads from the heater leads to prevent electromagnetic, 
electrostatic, and direct-emission effects. 

Furthermore, if the heater-cathode insulation is not very high and 
if there is impedance between the cathode circuit and ground, voltage 
at hum frequency will be introduced across this impedance into either 
the grid circuit or the plate circuit of the tube. This condition is 
made worse by the fact that it is common practice in amplifier designs 
to have a biasing resistor common to both grid circuit and plate cir- 

To show how these various factors are taken into account in cur- 
rent designs of vacuum tubes, the type RCA-1603 tube, illustrated in 
Fig. 1, may be taken as a typical example. From the standpoint of 
thermal agitation noise, regular vacuum tube design practice is such 
that no particular precautions have to be observed. Most common 
practice is to use either nickel or some alloy of nickel for both grid 
wires and side-rod supports. The choice of these materials is due to 
the high electrical conductivity of nickel and the fact that both are 
relatively easy to fabricate. 

From the standpoint of shot-effect noise, the cathode structure and 
heater are designed so that adequate emission will be obtained for all 
conditions of operation. The cathode has a capacity of electron 
emission several hundred times that ordinarily required for operation. 
A laboratory test for hiss is made of the tubes in a regular amplifier, 
and strict limits are maintained. In order to get low gas pressure and 
preclude any effects of ionization the temperature of the electrodes 
during the exhaustion process is controlled within very close limits. 
Micas in the tube that support and space the various electrodes are 
sprayed with an insulating compound that is chemically inactive even 
at high temperatures. This process not only reduces the likelihood 
of electrical charges building up on the mica, but also reduces the 
electrical leakage between electrodes. At some points in the micas, 
openings have been cut between the various supports to increase the 
leakage path. A still further precaution to reduce secondary-emis- 
sion effects is to coat the inside of the glass envelope of the tube with 
carbon. This reduces the possibility of unequal charges on the inner 


surface of the glass container. These unequal charges in trying to 
reach a state of equilibrium give rise to noise effects. Another pre- 
caution is to place the getter in such position that when it flashes none 
of the deposits will reach the stem press. The reason is that the 
material used in the getter, if allowed to deposit upon the stem be- 
tween the lead wires, would form a leakage path of variable resistance 
that would tend to make the tube noisy in operation. 

From the microphonic-response standpoint, certain structural 
features are incorporated. First of these is the use of nickel side-rod 
supports of large diameter, for rigidity and mechanical strength. 
The supporting micas are also made double the thickness used in con- 
ventional designs. In order further to reduce the possibility of elec- 
trode movement in the micas the diameter of the holes into which the 
grid and plate support rods are placed is from 0.0005 to 0.001 inch 
smaller than the diameter of the wires themselves. This insures a 
tightness and rigidity of mount that must be attained even at some 
expense in manufacturing ease and speed. The micas are then rigidly 
clamped in metal collars to strengthen the entire structure further. 
To center the entire mount within the tube envelope, four butterfly 
micas are used instead of the two customarily found in vacuum-tube 
practice. As pointed out previously, extra-heavy plate-support rods 
are used ; these are also suspended at three points of support instead 
of the customary two. Further to preclude mechanical movement 
of the heater within the cathode, the heater outside diameter, in- 
cluding the insulating coating, is held within 0.0005 inch of a specified 
diameter. This close tolerance results in a force-fit of the heater into 
the cathode and again illustrates a condition where performance is 
more important than convenience of manufacture. 

In order to reduce the hum to low orders of magnitude a double- 
helix type of heater winding is used. To insure that the heater-to- 
cathode leakage will be extremely small, very careful chemical con- 
trol is maintained over the ceramics used for spraying the heaters. 
Minute particles of iron, alkali, and other impurities may mean the 
difference between a usable product and one that is wholly unsuitable. 
In order to avoid any possible emission from the ends of the heater, 
centering the heater within the cathode sleeve is carefully supervised. 
To insure further that the heater is as nearly chemically pure as pos- 
sible, the wire is very carefully cleaned even before spraying it with in- 
sulating material. The cathode sleeve is also subjected to this same 
process for the same reason, Some control of the hum in the final 

44 L. C. HOLLANDS AND A. M. GLOVER [j. s. M. P. E. 

product can also be gained by specially aging the tube after exhaustion. 
Hum is also dependent to a large extent upon cathode temperature; 
the lower the operating temperature, the less the hum. Again, the 
design engineer, from experience, must balance between a low tem- 
perature and low hum, and a high temperature and low "shot" noise. 
Since the effects are complicated they can not readily be calculated 
mathematically and must be arrived at by cut-and-try methods. 
This brief analysis gives some idea of the problems confronting the 
tube engineer on special applications of vacuum tubes. 

Another typical example of special tube design is the type RCA- 
1609 tube, shown in Fig. 1, which is a filamentary type of sharp cut- 
off pentode. This tube finds its widest use in portable equipment at 
points remote from power lines, which must of necessity be operated 
on either dry batteries or storage batteries. In this tube extra-heavy 
mount supports are used to insure rigidity from the microphonic 
standpoint and also to withstand the rough handling to which portable 
equipment is likely to be subjected. The grids of the tube are sup- 
ported and spaced by glass beads instead of by mica, because it has 
been determined that glass beads are better suited mechanically to 
tubes for portable service. The filament is made of ribbon, in a short 
inverted V, for the same reason. Again, as in the 1603, the getter 
flash is directed in such a way that there is no tendency for deposits 
to form upon the stem press. The tube has a high emission limit for 
factory testing to insure freedom from shot effect. In order to in- 
sure low microphonic response, the tension of the filament during 
mounting is closely controlled. 

One of the newer tubes of this special group is the type RCA-1612. 
Its construction is similar to the standard type 6L7 tube in that it has 
two control grids separated from each other by a screen grid. The 
control grid nearest the plate is separated from the plate by another 
screen grid and a suppressor grid. Its electrical characteristics are 
similar to those of the standard 6L7. At the present time, its widest 
application is in a new amplifier design allowing remote control of 
volume without the usual difficulties experienced with circuits of this 
kind. Since the transconductance, and hence the gain, of the tube 
can be controlled by means of a d-c. bias applied to the control grid 
next to the cathode, the gain of a signal applied to the control grid 
nearer the plate may be controlled by the control grid nearer the 
cathode without the necessity of bringing the audio-frequency signal 
to the remote control point. It is thus possible to dispense with the 


elaborate shielding usually necessary for such an amplifier circuit. 
This opens up vast new possibilities for the motion picture engineer 
due to the extreme flexibility of the system. Other possibilities and 
design features will undoubtedly result from the availability of this 

Let us now consider phototubes and their applications in the field of 
motion pictures. The phototube is an essential part of modern 
motion picture projection equipment; by it the photographic record 
of music or speech is transferred back into audible sound. The design 
of a phototube is quite simple compared to that of a receiving tube, 
yet the procedure involved in producing the sensitive photo surface is 
complex and still not completely understood. 

The usual type of phototube contains a semicylindrical silver sur- 
face treated in such fashion as to become sensitive to light. The 
sensitizing process consists, in the main, in producing a layer of silver 
oxide upon which is deposited caesium; a reaction is then brought 
about, the final product being a mixture of caesium and silver and 
their oxides. This surface is the cathode or negative electrode of the 
phototube. Upon exposure to light of suitable color or wavelength, 
electrons are given off from the cathode in numbers proportional to the 
intensity of the light. If the output of a phototube is not linearly 
responsive to the light-intensity, the cause of the departure from 
linearity is to be found in some effect occurring subsequently to the 
emission of the electron current. 

The anode or positive electrode is a wire placed concentrically 
along the axis of the cathode, and may be made of any metal photo- 
electrically insensitive to light in the spectral region for which the 
phototube is designed. The modern caesium cell is particularly suit- 
able for use with incandescent light-sources because the major portion 
of its sensitivity lies in the red and infrared regions of the spectrum. 
The position of the anode serves to create a radial electrical field be- 
tween it and the cathode. This is important if the tube contains an 
inert gas that may be ionized and thus contribute to the photoelectric 
current that flows when the sensitive cathode is illuminated. The 
degree of ionization and consequent amplification is dependent upon 
the number of collisions between the photoelectrons and the mole- 
cules of the gas during the transit of the electron from cathode to 

The photoelectric tube is thus a diode, and possesses the rectifica- 
tion properties associated with such tubes. Current flows only when 


the anode is positive; thus, either a direct voltage or alternating 
voltage may be used for the anode, but in the latter case current will 
flow during half the cycle only. This fact may be useful in controlling 
the action of the tube in its associated circuits. It is important to 
consider the magnitude of the current that may flow under practical 
conditions. The most sensitive caesium surface will emit only 30 
to 50 microamperes per lumen of visible light (light-source at 2870K). 
When gas is used to amplify this primary current, the current is 
multiplied by a factor of five to ten, but at some sacrifice of perform- 
ance which will be discussed later. The illumination available will 
probably not exceed one lumen at best in sound equipment a few 
hundredths of a lumen is commonly used, so that the output current 
is to be measured in microamperes. 

The applications of phototubes may be divided roughly into two 
main classes: one in which the tube serves merely to trigger some 
mechanism such as a counter in this use there is no question of the 
nature of the photocurrent response other than that it be relatively 
instantaneous and positive ; and another in which the phototube is used 
to translate a rapidly varying photographic record into an accurate 
reproduction of the original sound. This record may contain fun- 
damentals and harmonics of a complicated pattern. The tube is re- 
quired to respond to frequencies in the neighborhood of 10,000 cps. 
and upward, and a flat frequency response is, therefore, desirable. 
Furthermore, the tube is part of equipment that must undergo con- 
siderable physical vibration, which would appear as "noise" in the 
sound output were the phototube not especially resistant to such 
causes of microphonics. An optimal lower limit to microphonic noise 
is reached only when the level of the noise is reduced below that of the 
noise introduced by the photographic film, which may be caused by 
the nature of the emulsion and its film base, and by noise inherent in 
the recording apparatus. By and large, the noise present for a given 
output of the so-called "gas-type phototube" is greater than that pres- 
ent in the output of a vacuum phototube; i. e., the signal-to-noise 
ratio is higher in the latter case. On the basis of this consideration 
alone, the vacuum type of phototube is to be preferred for sound use. 
However, with the improvement in sensitivity, a factor of five is cus- 
tomary, and the low output impedance of the gaseous type of tube has 
made its use popular in the motion picture industry. 

In this connection it is interesting to note that the use of a high im- 
pedance with a gas phototube reduces its effective voltage-sensitivity 

Jan., 1938 J 



to a level more nearly comparable to that of the vacuum type. For 
applications permitting the use of a high impedance, the vacuum type 
is preferable because of its high-fidelity frequency response and its 
improved signal-to-noise ratio. 

The effect of load impedance upon the voltage sensitivity of a gas 
phototube is shown in Fig. 2, which gives the plate families of curves 
for a gas and a vacuum phototube. Let us first consider the sensi- 


















> 09 C 





'/ y 


n 5 / 





\ A ' 














CE (R L X_ 


/ - 











45 M 

























__ . 









40 60 B' 80 8 100 


FIG. 2. Phototube anode characteristics. 

tivity of a phototube when used to actuate a relay; under such cir- 
cumstances the current that will flow, for a given amount of light, 
defines the static sensitivity S = I/L. When the tube is used to re- 
spond to modulated light, the sensitivity must be expressed as S d = 
dl/dL, a quantity that is akin to the transconductance of an amplifier 
tube. The phototube is usually used with a voltage amplifier so that 
a voltage sensitivity 5, = dV/dL = S d (R P R L )/(R P + R L ) is defined, 
again analogous to that of an amplifier tube. For a vacuum tube in 


L. C. HOLLANDS AND A. M. GLOVER [j. s. M. p. E. 

which R P is large, the voltage sensitivity reduces to S v = S d R L so 
that S v can be increased by increasing R L as long as the approximation 
holds. In Fig. 3 is shown the effective voltage obtained for a given 
light, from both gas and vacuum tubes, as a function of the load used. 
The radio tube owes many of its physical characteristics to its prede- 
cessor the electric lamp : the glass bulb and the nature of the stem 
press by which the electrical connections to the interior elements are 
made are but two of these inherited traits. For a tube as simple elec- 
trically as the phototube the cumbersome stem press is unnecessary ; 

20 40 60 80 


FIG. 3. Phototube operating characteristics. 

furthermore a considerable part of the electrical capacity present, 
which may limit the response of the tube at high frequencies, is to be 
found between the lead wires of the stem press. Hence it was decided 
to introduce a line of phototubes mounted in glass tubing and analo- 
gous to the Lum-i-line lamp in construction. These tubes are the 
RCA-921 and RCA-922. The cathode and anode are mounted inde- 
pendently of each other at opposite ends on chrome-iron caps in- 
sulated from each other by a length of glass tubing indeed, in the 
vacuum-type tube of this construction one of the metal caps serves as 
the anode, the concentric wire being found unnecessary when an 


ionizing gas is omitted. By this construction the electrical capacity 
has been reduced to 1.1 /ijuf and 0.6/xMf, for the gas and vacuum types, 
respectively, as compared with 2.4 /z/*f and 2.2 ju/*f for the correspond- 
ing bulb types. The smaller cathode is of course responsible for part 
of this reduction. The tubes have been designed with the point in 
mind that the user may wish to make direct connection from one 
contact of the phototube to the grid of the first amplifier tube, for 
which purpose an amplifier with grid top connection is desirable. 
The phototube may be supported on one end, by a simple contact as 
compared with the customary tube socket. To orient the tube with 
respect to the direction of the light-beam, a square metal clip has been 
added which serves also to protect the exhaust tubing. The compact 
construction thus resulting should prove particularly adapted for use 
with 16-mm. equipment where economy of space 
is a desirable feature. The RCA-922 (the RCA- 
921 is similar structurally) is shown in Fig. 4. Its 
height is about 2 inches over all. 

The sensitive surface of a phototube may change 
considerably during the life of the tube. When 
a surface is first prepared, a high sensitivity is 
obtained which gradually decreases with use until 
a steady value is reached. The aging schedule 
adopted for the production of phototubes is based 
upon this initial decrease of sensitivity. It is 
found that a large fraction of the drop may be 
covered during the aging schedule, so that only $#? tube. 

a small further decrease may be expected during 
use. When a tube stands for any length of time, the sensitivity 
may increase temporarily to a value close to the original high sen- 
sitivity ; for that reason it is recommended that a phototube that is in 
use only intermittently be operated for some length of time before 
being put into service. 

The drop with use is much more pronounced in gas tubes than in 
vacuum tubes. Little attention need be paid to a vacuum tube; 
its characteristics remain unchanged over long periods of time. 

In summing up the applications of the phototube to sound pic- 
tures mention may be made of the electron multiplier, which con- 
sists of a photosensitive surface and a number of secondary emitting 
surfaces so arranged that the primary photoelectron current may be 
amplified by large factors depending upon the number of secondary 


L. C. HOLLANDS AND A. M. GLOVER [J. s. M. p. E. 






, 420 2 


(.4 ) Condensers C and C e have been chosen to give output voltages equal to 
0.8 Eo for /i of 100 cycles. For any other value of f\, multiply values of C and C e 
by 100//1. 

In the case of condenser C c , the values shown are for an amplifier with d-c. 
heater excitation. When a-c. is used, depending upon the character of the as- 
sociated circuits, the gam, and the value of /i, it may be necessary to increase the 
value of C c to minimize hum disturbances. It may also be desirable to have a d-c. 
potential difference of approximately 10 volts between heater and cathode. 

(B) /a = frequency at which high-frequency response begins to fall off. 

(C) The voltage output at/i for n like stages equals (0.8 <>)". 

(D) Decoupling filters are not necessary for two stages or less. 

(E) For an amplifier of typical construction, the value of ft is well above the 
audio-frequency range for any value of RL- 

(F) Always use highest permissible value of R g . 

(G) A variation of = fc 10% in values of resistors and condensers has only a 
slight effect upon performance. 




Co e 


f, 420 ~ fi 


The diagram given above is for Phase-Inverter Service. 'Phe signal input is 
supplied to the grid of the left-hand triode unit. The grid of the right-hand unit 
obtains its signal from a tap P on the grid resistor R in the output circuit of the 
left-hand triode unit. The tap P is chosen so as to make the voltage output of 
the right-hand unit equal to that of the left-hand unit. Its location is deter- 
mined from the voltage gain values. For example, if the voltage gain is 20, P 
is chosen so as to supply 1 / 20 of the voltage across R g to the grid of the right-hand 

For phase-inverter service, the cathode resistor R e should not be by-passed by 
a condenser. Omission of the condenser in this service assists in balancing the 
output voltages. The value of R c is specified on the basis that both units are 
opera ting simultaneously at the same values of plate load and plate voltage. 

FIG. 5 ( Upper). 

Single-stage resistance-coupled triode amplifier. 
Resistance-coupled twin-triode amplifier. 


surfaces used. If gains of 3.5 or 4 are obtained at each surface the 
overall gain for various numbers of stages is as shown in Table I. 

The electron multiplier is a current amplifier; the power output is 
limited by the nature of the secondary-emission surfaces. The out- 
put characteristic is quite similar to that of a vacuum phototube; 
however, the plate voltage is measured with respect to the last sec- 
ondary-emission surface which itself may be several hundred volts 
above the photocathode. At least one hundred volts per stage of 
secondary emission is required to obtain a sufficient ratio of secondary 
to primary current. These tubes should not be used in commercial 
designs since they are developmental tubes and the characteristics 
are still undergoing changes of a design nature. 


Gains for Various Numbers of Stages in Electron Multiplier 

Stages Gain 

1 3.5- 4 

2 12.2- 16 

3 43-64 

4 150 - 256 

5 530 - 1,024 

6 1,850 - 4,096 

7 6,500 -16,384 

8 20,000 -65,536 

Fig. 5 contains data that will be data of assistance to engineers in 
the motion picture field who are interested in designing resistance- 
coupled amplifiers. 

It is well to remind the design engineer at this point that, in gen- 
eral, conservative design produces the most economical apparatus 
from the standpoint of life. This is particularly true of vacuum-tube 
apparatus where long life and uninterrupted service are prime con- 
siderations. In such applications, it is very essential to design so 
that tubes are operated under moderate voltage and load conditions. 

Difficulties from low-frequency oscillation (motorboating) are 
often experienced when a high-gain multistage audio amplifier of con- 
ventional design obtains its B voltage from a single power-supply 
unit. These difficulties are usually due to interstage coupling through 
a common impedance in the power unit. Customary expedients to 
prevent motorboating include the use of very large filter condensers 
and several power-supply units. However, by suitable design, the 




^. g 2 M g p; S 


" ^ R S . **. S . t 
o - o * o 

OR ( 

AIN 4 




5 S ; 


5 15- 

d d " 

5f S 

tu o 

5 5 


gain of an audio amplifier of low frequencies can be reduced to such 
levels that the effects of feed-back through a power unit of conven- 
tional design are greatly reduced. 

The use of a series screen resistor and a self -bias resistor offers sev- 
eral advantages over fixed- voltage operation: (1) the effects of pos- 
sible tube differences are minimized ; (2) operation over a wide range 
of plate-supply voltages without appreciable change of gain is fea- 
sible; and (3) the low frequency at which the amplifier cuts off is 
easily changed. Fixed-bias or fixed-screen voltage operation in- 
creases the tendency of an amplifier to motorboat and decreases the 
compensating action of the remaining series resistors. The advan- 
tages of an amplifier constructed according to the data presented 
herewith can be further emphasized by the addition of suitable de- 
coupling resistors and condensers. With a proper decoupling filter 
in the plate circuit of each stage, three or more amplifier stages can be 
operated from a single power-supply unit of conventional design 
without encountering difficulties due to coupling through the power 
unit; not more than two stages should be operated from a single 
power-supply unit when decoupling filters are not used. 

Detailed information on the operation of the 6C6, 6J7, and 57 as re- 
sistance-coupled audio-frequency amplifiers is presented in Fig. 6. 
These data hold for plate-supply voltages from 90 to 600 volts, for 
plate-resistor values of 0.1, 0.25, and 0.5 megohm, and for a number of 
grid-resistor values. The combination of resistor and condenser values 
suggested in the pentode chart causes a 30 per cent drop in output 
voltage per stage at 100 cps. A similar cut-off characteristic at any 
other low frequency (/ ? ) can be obtained by multiplying the capaci- 
tance values shown in the chart by 100// z . 

Detailed information on the operation of triodes as resistance- 
coupled audio-frequency amplifiers is presented in Fig. 7. The 
combination of condenser and resistor values suggested in this chart 
causes a 20 per cent drop in output voltage per stage at 100 cps. 
A similar cut-off characteristic at any other low frequency (f t ) can be 
obtained by multiplying the capacitance values shown by 100// f . 
As with pentodes, the use of self-bias reduces the effects of possible 
tube differences and permits operation over a wide range of plate- 
supply voltages without appreciable change of gain. The regulating 
action of a self -biased triode amplifier is not as good as that of a pen- 
tode amplifier having series screen and self-biasing resistors, because 
the regulating action of a screen is not available in a triode. 


L. C. HOLLANDS AND A. M. GLOVER [j. s. M. p. E. 


" ; 


s ? 



i! j: 
1 2 



-i?.iit S 

5 1 2 * 

15* = !! = 




! = = = 



I 68 

- 8 S 

rf i 5 

a u 5 




When a number of high-mu triode amplifier stages are cascaded, 
the high-frequency response may be severely curtailed, because the 
high effective input capacitance of a triode shunts the load of the 
previous tube. When good high-frequency response from a triode 
amplifier is desired, therefore, low-mu tubes and low values of plate 
and grid resistors should be used. 

On the charts, the values of C c are given for an amplifier with d-c. 
heater excitation. When alternating current is used, depending 
upon the character of the associated circuits, the gain, and the value 
of f lt it may be necessary to increase the value of C c to minimize 
hum disturbances. It may also be desirable to have a d-c. potential 
difference of approximately 10 volts between heater and cathode. 


MR. CRABTREE: What metals other than nickel are used in vacuum tubes? 

MR. HOLLANDS: Molybdenum, copper, some iron; practically all metals 
normally known to ordinary manufacturing technic. The plates and side-rods 
of the tubes shown here are of nickel and the grids are of molybdenum nickel, or 
what we call "moly nickel." The filament is thoriated tungsten in some cases; 
in most receiving tubes, oxide-coated nickel wire or cathodes are used. Not much 
tantalum is used in receiving tubes, but more in transmitting tubes. 

MR. TUPPER : Is there any difference in materials between a regular RCA tube 
and a tube licensed by RCA? 

MR. HOLLANDS: I can not answer that question, because our licensees use any 
metals they find desirable in the construction of tubes. Ordinarily they use about 
the same materials that we do. 

MR. CRABTREE: Does RCA insist that the manufacturers of tubes under its 
licenses maintain the standards set by RCA? 

MR. HOLLANDS: No. However, progress is being made in standardizing the 
electrical characteristics of receiving tubes by the Radio Manufacturing Associa- 
tion. The physical structure or the materials used in the tubes are entirely up to 
the licensees. 

MR. PALMER: Can the sounds from a photocell be heard through a pair of 
head-phones without amplification? If not, how many stages are needed? 

MR. HOLLANDS: The output of the phototube is measured in microamperes, 
so it would be necessary to use one or two stages of amplification. 

MR. FARNHAM: I have heard designers of high-grade amplifiers, particularly 
those used for picture transmission, say that the conventional tubes supplied 
for broadcast receivers are not of sufficiently high quality, and that special tubes 
were desirable. Are there two grades of tubes? 

MR. HOLLANDS: Yes; we make a special line of tubes called the 1600 series, 
in which we go to extreme lengths to assure mechanical rigidity, perfect spacing, 
and other important features. Those tubes are more expensive, because their 
manufacture is very much slower and they require considerably closer supervision 
in production. They also go through several tests that are not given to regular 

56 L. C. HOLLANDS AND A. M. GLOVER [J. s. M. P. E. 

amplifier tubes. For instance, they are tested directly under actual voltage 
amplification conditions. 

MR. FARNHAM: Does the sensitivity of a phototube depend upon the width of 
the spectral band impinging upon it? 

MR. GLOVER: The caesium tube, which has a high microampere-per-lumen 
sensitivity, is very sensitive to the red and infrared so that if used with a mono- 
chromatic light-source in the blue, it will show a much lower sensitivity. In the 
blue region a potassium surface would be much more sensitive. 

MR. FARNHAM: And if all the illumination were concentrated in the blue or 

MR. GLOVER: Then the caesium tube would be much less sensitive than the 
potassium tube. 

MR. KELLOGG: In our sound reproducing optical system no measures have 
been taken to exclude the infrared light. Is it likely that the infrared is con- 
tributing very largely to the response? What per cent of the light will an infrared 
filter exclude? 

MR. GLOVER: Quite a large fraction. The maximum sensitivity of a caesium 
tube occurs between 7000 and 8000 A, which is at the very edge of the visible 
spectrum. The sensitivity extends to 12,000 A, and over that range the emission 
from a tungsten lamp rises rapidly, so that the response of the tube in that range 
is, roughly, perhaps 60 per cent of the overall response. 

MR. MALMUTH: I understand that these tubes are mounted directly on the 
first-stage amplifier tube. 

MR. HOLLANDS: Not directly upon it. They can be mounted very close to 
it, so that the small cap is at the top and relatively close to the grid-cap of the 
amplifier tube. The grid-cap of the amplifier tube is round, not square, like the 
chrome cap of the phototube. The reason for making the latter square was to 
permit orienting the tube with regard to the light-beam. 

MR. PALMER: It is customary to rate exciter lamps in volts and amperes, 
say, 4 volts, 0.75 ampere. Would we have as much photoelectric effect if we 
operated them at 2 volts? The life would be much longer. 

MR. GLOVER: As the voltage of an incandescent lamp is raised, a greater and 
greater fraction of the light emitted falls within the visible spectrum. At very 
low voltages only heat is given off, which means that the energy radiated is in the 
far infrared, which is beyond the sensitivity of the tube. 

MR. PALMER: I did not mean to go that far. I meant some compromise 
between the rated voltage and the voltage necessary to activate the photocell. 

MR. GLOVER: You will find a considerable decrease if you attempt to cut the 
voltage to any extent. 

MR. FARNHAM: Reproducer optical systems pick up only a small area of the 
light-source, and if you reduce the brightness or the color-temperature of the 
exciter lamp, then the light that is available is less, and more amplification must 
be used. As a result, other troubles may enter due to, say, ground-noise ; whereas, 
if the lamp is operated at a high order of brightness, a great deal more light will 
pass through the optical system. It is a little different from using a lamp without 
an optical system; if the entire luminous output of the filament were available 
you could substitute wattage for brightness and get the same effect and thus 
longer life. 


MR. WEISS: How do the 922 and 921 tubes compare in characteristics with the 
present phototubes, for instance, the 868? 

MR. GLOVER: The 922 is electrically identical to the 917. The gas type is 
very nearly the same as the 918. The 921 and the 918 have higher sensitivity 
than the 868. 

MR. KELLOGG : Photoelectric tubes operate saturated, and there is no appreci- 
able space-charge effect. Do not practically all microphonic effects disappear 
under such conditions? 

MR. HOLLANDS: No. There are several microphonic effects that are rather 
difficult to trace, but we have found that the microphonic troubles we have at the 
present time are due almost entirely to physical movement of the elements within 
the tube. That is easily demonstrated with a Strobotac, by vibrating the tube and 
using any two elements of it as a condenser microphone, "stopping" the movement 
of the two elements with the Strobotac, and watching the amplitudes of the output 
amplifier. Relatively few microphonics are due to electrical characteristics of 
the tube. 

MR. KELLOGG: I do not believe I made the question clear. Assuming there 
is movement of the electrodes with respect to each other, it is hard to imagine 
that that will alter the current through the device except through the agency of 

MR. HOLLANDS: That is right. 

MR. KELLOGG: Working so close to the saturation point, then, would there be 
enough space-charge effect to cause microphonic noise in the circuit? Our 
experience has been that photocells are relatively immune from microphonic 

MR. GLOVER: There is very little trace of microphonics in a vacuum photo- 
tube. We have recently measured the microphonics in the gas type of tube, and 
find that they rise rapidly with increase of voltage and increase of gas amplifica- 
tion. Apparently the cause lies in the gas, although at present we can give no 
explanation for it. 

MR. CRABTREE: Is the geometry of the tube arrived at by trial-and-error, or 
do you apply a mixture of trial-and-error and mathematical methods? 

MR. HOLLANDS: The tube design engineer has an idea of the characteristics 
for which he is striving. From curves and mathematical data that we have, we 
can derive a pretty close first approximation by calculation. The factory then 
fabricates the tubes, and minor changes are made to produce the desired charac- 
teristics. Briefly, we arrive at the first approximation mathematically, and then 
by cut-and-try methods we trim the characteristics. 




Summary. The luminous and electrical characteristics of a number of vapor 
discharge lamps that have attained practical importance in recent years are described. 
These include the sodium vapor lamp, the high-intensity mercury vapor lamp, and 
the high-pressure quartz capillary lamp. The fundamental physical phenomena 
and the manner in which these affect the light output and efficiency are discussed 
briefly. The effect of variations in gas pressure and current density upon the dis- 
tribution of intensity in the spectrum is dealt with, and also the accompanying changes 
in intrinsic brilliancy and color of light emitted. 

The latter part of the paper contains a discussion of recent developments in the 
utilization of fluorescent materials in gaseous discharge lamps. These lamps offer 
interesting possibilities from the point of view of general illumination and special 
color effects. 

During the past few years there has been considerable development 
in the utilization of electric discharges in gases and vapors for the 
production of light at higher efficiencies than those obtainable with 
incandescent lamps of similar wattage. It is the object of the follow- 
ing remarks to describe some of these new light-sources from the point 
of view of the engineer who is interested in their application. 

As a result of the knowledge that has been gained during the past 
two decades on the nature of the emission spectra of different ele- 
ments and compounds, we find that the only discharges that are of 
practical interest as light-sources are those in neon, mercury vapor, 
and sodium vapor. Logically, we should include under the same 
general heading arcs between carbon or impregnated carbon elec- 
trodes in air at ordinary pressure. However, since the phenomena in 
such arcs are different in certain respects from those observed with 
discharges in the monatomic vapor, they will not be discussed in the 
present connection. 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received October 
6, 1937. (A summary, with slight additional material, of a paper published in 
J. Opt. Soc. Amer., 27 (Jan., 1937), No. 1, p. 1, where a comprehensive bib- 
liography is given.) 

** Research Laboratory, General Electric Co., Schenectady, N. Y. 


Luminous Efficiency. In any consideration of light-sources, it is 
first of all necessary to take into account the characteristics of the 
human eye as a detector of radiation. What is usually designated as 
"light" is actually a rather narrow region in the extremely wide range 
of electromagnetic radiations that have been observed, and are shown 
in Table I. 


Lower Range of Corresponding 

Wavelength in Elec- 

Radiation Angstrom Units tron Volts 

Gamma Rays 0.1 123,360 

X-rays 1.0 12,336 

Far Ultraviolet 1000 12.34 

Near Ultraviolet 3000 4.11 

Visible 4100 3.01 

Infrared 7200 1.71 

Shortest Hertzian Waves 10 7 = 0.1 cm. 10~ 3 

"Radio" Waves 1 meter 10 ~ 6 

1 kilometer 10 ~ 9 

The second column gives the value in Angstrom units (1 A = 10 ~ 8 
cm.) of approximately the shortest wavelength in the particular 
region. Since there are no sharp dividing lines between these regions, 
these values are to be regarded as rough indicators of the extent of 
each type of radiation. The third column gives the voltage through 
which an electron would have to be accelerated to cause the emission 
of the given wavelength. As will be pointed out subsequently, there 
is a definite relation between the two magnitudes, which is expressed 
by the relation 


where X is the wavelength in Angstrom units, and V is measured in 

Fig. 1 shows the average visibility curve for the eye. For all practi- 
cal purposes the visibility V x may be taken as zero for all wavelengths 
outside the region included between Xi = 4100 A and X2 = 7200 A, 
and has a maximum value for the radiation in the green which has a 
wavelength X = 5550 A. A source of light emitting radiation of this 
wavelength exclusively would have an efficiency of 621 lumens per 
watt. On the other hand, the most that we could expect from a 
source emitting only the characteristic yellow radiation of sodium 



fj. S. M. P. E. 

vapor (X = 5890, 5896) is 475 lumens per watt, and for red or blue 
the optimum efficiency would be much less. 

While a tungsten filament lamp or other incandescent solid emits 
radiations of which the wavelengths vary continuously from the ultra- 
violet into the infrared, the radiation from a monatomic gas or vapor 
such as neon or mercury, when examined by a spectroscope, is found 
to consist of a number of distinct lines, each corresponding to a 
definite wavelength. A few typical line spectra are shown at the top 

' V 1 B G ' K ' O ' R~~ Infj-Q Red *" 

\ 2300 *C{OQ 

*500 +<pCO 4500 1000 6006 6OOO lOOOO *OOO ZOOC 






III 1 1 

"9 II 1111 


Color Distinctive Visibility r\ 
L/nes f \ 


Vio/et 4zoc 

' 0. OlZ / \ 

Wve rfq 4358 . 0175 1 \ * 
He 4471 .033 1 \ 


Green Hq 4960 

.27 \ 

/Va 4978 

498^ .30 / \ 

No 3/34 

'" \ ^~ 


Hq 5461 

1 ^ 

YellOW j 5770^5791 .89 / \ ^ - 

N6 58S& 
He 5676 

.81 ^ 
.78 \ ^ - 


Ha 5890,5896 .786 / I 

Oronoe Na e/eo .43 1 \ 

7 Ne 640Z .173 1 \ 


Red He 6678 

.039 / \ 


.016 / \ 

A ZSOO 3000 

33OO 4000 ^/ 50pO 6000 \^ 8OCO lOOOO 130QO &poq 

FIG. 1. Some typical line spectra and visibility curve. 

of Fig. 1, while the table in the left-hand corner gives the wave- 
lengths, corresponding colors, and visibilities (in terms of the maxi- 
mum visibility as 1). It is the predominance in intensity of one or 
more of these lines that gives rise to the characteristic colors of the 
light obtained from these gases and vapors. 

On the basis of the observed distribution of energy in the spectrum 
of any light-source and the visibility curve shown in Fig. 1, it is 
possible to calculate a theoretically optimum efficiency, which we 
shall designate by L . Table II gives values of L for different 
sources and, for comparison, values of efficiency L s actually obtained 
on practical lamps. The third column gives the percentage 


of the total energy emitted by the source that is in the visible range. 
This is deduced by means of the relation 

r, = L./L 

The last column gives the energy utilization ratio e, that is, the ratio 
between the total watts input and the watts emitted as visible light 
(which is measured in terms of lumens). 


Luminous Efficiencies of Various Sources of Light 
(Lumens per Watt) 

Source L L, 100 100 

Black Body at T = 6500 218 86.3 39.5 13.9 

Sun 250 100 40 16.1 

Tungsten (Gas-Filled) 143 15-30 10 -20 2.5-5.0 

Flaming Arc 220 45-75 20 -34 7.2-12.1 

Sodium Vapor 475 50-75 10 -15 8 -12 
Mercury Vapor 

Low-Pressure 248 15-20 6 -8 2.5-3.2 

1 Atm. (Type //) 298 30-35 10 -12 4.8-5.6 

Higher Pressures 298 40-50 13 -17 6.4-8.0 

Neon 198 15-40 7.5-20 2.5- 6.4 

Helium 4-10 

Carbon Dioxide 2- 4 

Cadmium 0.5-1 
Green Fluorescent 

(L. P. Mercury) 475 60-80 12 . 6-16 . 9 9 . 6-12 . 9 

Physical Phenomena in a Gaseous Discharge Lamp. In an incan- 
descent solid the light is produced as a result of the high temperature 
to which the conductor is raised by the passage of the current. The 
light output increases with the temperature, and tungsten has be- 
come the metal for use in incandescent lamps because it has a higher 
melting point than any other metal. 

On the other hand, the mechanism by which light is produced in a 
gaseous discharge is quite different, and we must therefore consider 
briefly the nature of this mechanism. This may be discussed under 
two headings: (1) the fundamental processes by which an atom or 
molecule of any gas or vapor may be made to emit radiation, and (2) 
the processes of electrical conduction in a gas or vapor. Primarily, a 
gaseous discharge lamp is a conductor of current, and the essential 
characteristics of the conduction phenomena are the same whether 
the gas used is argon or neon. But in the case of the former there is 

62 S. DUSHMAN [J. S. M. P. E. 

very little visible light, since most of the lines in the spectrum of 
argon occur in the ultraviolet and infrared regions, while in the case 
of neon, as evidenced by its practical utilization, the spectrum is rich 
in lines in the red end. 

Our present views on the origin of spectral lines are based upon a 
theory that was first postulated by N. Bohr in 1913, and has since 
then been found to be in excellent agreement with observations by a 
large number of investigators who have worked in this field during the 
past twenty-four years. These views may be understood best by 
describing briefly an experiment that was carried out by two German 
physicists, J. Franck and G. Hertz, in 1915, in order to test certain 
deductions from Bohr's theory. 

It is well known that when a heated tungsten filament is used as a 
cathode (negative electrode) in a highly evacuated bulb, electrons 
are emitted. These carry the current to the anode (positive elec- 
trode, or "plate"), and the magnitude of the current thus transported 
by the electrons depends upon both the temperature of the cathode 
and the plate voltage. The kinetic energy of the electrons is propor- 
tional to the anode voltage in accordance with the relation 

V 2 wz' 2 = Ve 

where e = charge on electron, 
m = mass of electron, 
v = velocity of electron, 
V = positive potential on anode. 

Such an evacuated device is used as a rectifier in radio sets because 
the electrons can pass in only one direction, as long as the plate is not 
at a sufficiently high temperature to emit electrons at any consider- 
able rate. Now, into this bulb we insert a pellet of metallic sodium 
and increase the temperature of the walls so that the vapor pressure 
of the sodium reaches a value of about one-millionth of an atmosphere. 
(This is the order of magnitude of the pressure in a 10,000-lumen 
sodium vapor lamp.) With the tungsten cathode heated to a tem- 
perature at which electrons are emitted in considerable numbers, we 
gradually increase the anode voltage from zero. At first, as long as 
this is below 2.1 volts, nothing happens; but at 2.1 volts, or 0.1 volt 
higher, it is observed that the vapor emits a spectrum consisting of 
only the two ZMines of wavelengths 5890 and 5896 A. These are the 
most prominent lines in the spectrum of sodium, and impart to the 
light its yellow-orange color. 

Jan., 1938] 


As the voltage is increased still higher, more lines appear in the 
spectrum, until, at 5.1 volts or higher, the whole spectrum is emitted. 
Now, what happened in this experiment is to be interpreted thus : 

FIG. 2. Energy levels and lines in arc spectrum of sodium. 

When an electron possessing the velocity corresponding to an accelera- 
tion through 2.1 volts collides with a sodium atom, the latter is 
excited to a higher-energy state, and when the system passes spon- 
taneously from this higher state to the normal, monochromatic 
radiation is emitted in accordance with the relation 



hv = Ve, 

[J. S. M. P. E. 

where v is the frequency of the radiation and h is a universal constant 
(known as the quantum constant). The corresponding wavelength is 
given by the relation 


where c is the velocity of light. 


'So 'P, 'Dz 

3 D 3 


FIG. 3. Energy levels and lines in arc spectrum of mercury. 

Each line in the spectrum of sodium corresponds to a similar transi- 
tion between a higher- and lower-energy state of the sodium atom, 
and each of these states requires a definite electron energy (corre- 
sponding to a so-called critical potential) for its excitation. Fig. 2 

Jan., 1938] 



shows these so-called "energy-levels" in the spectrum of sodium, and 
some of the lines that are observed spectroscopically. Fig. 3 is a 
similar energy -level diagram for mercury. It will be observed that 
while in the case of sodium the first excited state occurs at 2.1 volts, 
the lowest excited state for mercury occurs at 4.9 volts. Conse- 
quently, the corresponding transition in the latter case gives rise to a 
line of shorter wavelength (X = 2537 A), which is in the ultraviolet. 
Lines, such as the latter, corresponding to transitions between the first 
excited state and the normal state of an atom are known as resonance 
lines, since the atoms also absorb these radiations. Furthermore, the 

1.00 - 

2.0 2.2 2.4 2.6 2.8 

FIG. 4. Visibility as function of electron volts. 

excited state from which a transition can occur with emission of 
resonance radiation is designated a resonance state (or level). In 
order to obtain visible light from mercury it is necessary to excite the 
mercury atom to about 6.7 volts or higher, and the whole spectrum 
appears only when the electron velocities exceed that corresponding 
to 10.4 volts. 

Similarly, in the case of other spectra, the production of any line 
requires that the electron shall acquire a minimum value of the kinetic 
energy. Obviously, the most efficient light is obtained when the 
kinetic energy of the electron can be converted completely into visible 
radiation. Now, the visible radiation, as stated previously, extends 
from about 4000 to 7000 A. To produce these radiations the mini- 
mum values of the electronic energy must lie between about 1.8 

66 S. DUSHMAN [J. S. M. P. E. 

and 3.0 volts. Fig. 4 shows the visibility curve plotted against 
electron volts instead of against wavelength, as in Fig. 1 . The maxi- 
mum visibility is obtained by collisions of atoms with electrons of 
about 2.25 volts. 

When we examine the energy-level diagrams of the three elements, 
sodium, mercury, and neon, it is seen that only in the case of sodium 
is it possible to convert the energy of 2.1-volt electrons into light. 
On the other hand, in the case of the other two elements, the lowest 
excited levels are so high that the only radiation that can arise as a 
result of transitions from these levels to the normal lies in the ultra- 
violet. To obtain visible light the atoms have to be excited to above 
6.7 volts in the case of mercury and to above 19 volts in that of neon. 
Thus it would appear that at the maximum we could not expect an 
efficiency of light production from neon greater than about 2.2/19 = 
1/9 of the optimum ; and in the case of mercury similar considerations 
lead to the expectation that the maximum efficiency would be only 
about 2/8 of the optimum. 

Actually, as Table II shows, the values of L s , the specific luminous 
efficiencies observed for neon and mercury, are greater than predicted 
by this simple argument. While it is not practical in the present 
discussion to consider in detail the reasons for these observations, 
some of these may be mentioned briefly. 

First, even when the electrons possess the requisite energy to excite 
an atom by collision, this does not take place at every such collision. 
There exists a definite probability of excitation, which varies with the 
nature of the atom and the speed of the electron. 

Second, at higher pressures and higher current-densities an excited 
atom may suffer a collision with either another electron or another 
atom before a transition can occur that is accompanied by emission 
of light. 

Third, in order that the electrons may be able to pass freely from 
the cathode to the anode of the discharge, it is necessary to have pres- 
ent a minimum concentration of positive ions to neutralize the nega- 
tive space-charge otherwise produced by electrons. In order to 
produce these ions the electrons have to acquire a velocity corre- 
sponding to the ionization potential, and in a gas discharge this gives 
rise to a voltage drop at the cathode that is considerably greater than 
the voltage required for the excitation of visible light. 

The magnitude of this cathode drop depends upon the electron 
emissivity of the cathode, and is lower the higher the electron emis- 


sivity. At a cold cathode, such as those used in ordinary Geissler 
tubes, the electrons are pulled out, as it were, from the cathode by the 
intense electric field produced there by a high concentration of posi- 
tive ions. As a result, the cathode drop may vary from slightly less 
than 100 to 300 volts or higher, depending upon the nature of both 
the cathode and the gas. 

On the other hand, in the case of a thermionic cathode such as an 
oxide-coated nickel electrode, which emits electrons in virtue of its 
high temperature, the cathode fall ordinarily does not greatly exceed 
the ionization potential of the gas. Since this has the value of 25.5 
volts for helium, 21.5 volts for neon, 10.4 volts for mercury, and 5.1 
volts for sodium, and a fraction of a volt in excess of the ionization 
potential is required to produce all the ions necessary for conduction 
of the electrons through the gas, the use of hot cathodes in gaseous 
discharges has made it possible to operate such discharges at much 
lower voltages and higher currents than in the case of Geissler dis- 

Summarizing the foregoing remarks, it is important to recognize 
that any gaseous discharge lamp is fundamentally a conductor in 
which more than 99 per cent of the current is carried by electrons 
passing from the cathode to the anode, and the small residual current 
is carried by positive ions moving in the opposite direction. The 
light emitted is merely a by-product of the electrical phenomena in 
the discharge. The fundamental processes consist in the collision of 
electrons with the atoms of the gas or vapor. These collisions result 
in the formation of excited atoms and ions. The latter serve to 
eliminate electron space-charge at the cathode, and thus make it 
possible for the discharge to operate at appreciable current-densities. 
The nature of the atoms governs the type of spectrum emitted, and 
therefore affects the intensity and color of the light emitted, as well 
as the efficiency of light production. 

In recent years many investigators in different laboratories, both in 
this country and abroad, have studied these phenomena of conduction 
and light production in gases, and as a result there have been devel- 
oped a number of gaseous discharge lamps that have found consider- 
able commercial application. Not only are most of these light- 
sources more efficient than gas-filled tungsten lamps of similar 
wattage, but they possess other characteristics that should be of 
advantage in many special applications. In the following paragraphs 
some of these gaseous discharge lamps will be described briefly, and 

68 S. DUSHMAN [J. S. M. P. E. 

mention will be made of those characteristics that may be of interest 
in the present connection. 

Types of Gaseous Discharge Lamp. Since in the following discus- 
sion we shall consider only discharge lamps in which hot cathodes are 
used, the classification suggested by C. G. Found 2 will be adopted. 
Found classifies them, according to the geometry of the containing 
vessel, into two types: (1) cathodic, (2) positive column; and defines 
these as follows : 

"A cathodic discharge is defined as one that is more or less bulbular in shape 
and in which the distance between cathode and anode is comparable with the 
smallest dimension of the bulb. 

"A positive column is an elongated tube in which the distance between cathode 
and anode is several times the diameter." 

In general, it may be stated that a cathodic type of discharge oper- 
ates with a voltage drop that is approximately equal to the ionization 
potential of the gas, and may be as low as the resonance (lowest 
critical) potential (owing to successive impacts). A tungar rectifier 
is an example of such a discharge, and the d-c. low-voltage sodium 
lamp is another example. Such an arc may be started on a compara- 
tively low- voltage circuit (less than 110 volts) without any auxiliary 
voltage "kick." 

In a positive-column discharge the total voltage drop usually ex- 
ceeds the ionization potential for the gas, and while the discharge 
tube may be designed to operate on less than 110 volts, a starting 
kick or some equivalent device, such as an auxiliary electrode, is neces- 
sary. This initial high voltage is needed to overcome the negative 
charge on the walls tending to prevent the flow of electrons. The 
low-pressure mercury-vapor lamp with mercury cathode and the 
hot-cathode high-current neon tubes are examples of this type of dis- 
charge. The positive -column sodium- vapor neon lamp and high- 
lumen output a-c. sodium lamp developed for use on 6.6-ampere 
constant-current circuits (described in a subsequent section) are other 
examples. The main distinction between the two forms of discharge 
is in the fact that in the cathodic type all the energy from the external 
source of supply is converted into kinetic energy of electrons at the 
cathode, and, consequently, there is no voltage gradient outside the 
region of the cathode fall, as in a positive column. In the latter, 
there is, in addition to the cathode fall, a drop in the rest of the tube 
that varies with the length, other conditions remaining constant. 
In a cathodic discharge the extent of the light region is governed by 


the distance that the electrons traverse before they lose so much 
energy that they can not excite any atoms by collision. Such elec- 
trons are known as "ultimate" electrons in contrast to the "primary" 
electrons, which have acquired high kinetic energy in passing through 
the region of cathode fall. The light output in such a discharge is 
therefore a function of both the gas pressure and the cathode fall, and 
the latter is in turn governed by the thermionic emissivity of the 
cathode. As has been emphasized already, the function of the ions 
is merely that of eliminating space-charge. In consequence, the 
voltage drop in the arc adjusts itself to such a value for any given 
arc current that a sufficiently strong field will exist at the cathode 
both to provide the necessary electron emission and to enable the 
electrons to acquire sufficient energy to produce the necessary amount 
of ionization. 

In a cathodic discharge in neon-sodium the pressure of neon is 
about 1 mm., and under conditions of optimum light output the pres- 
sure of sodium vapor is about 0.001 mm. The value of the arc drop 
actually observed in such a discharge is about 18 or 19 volts when 
optimum light output is obtained, and it is only when the vapor pres- 
sure of sodium increases to values above 0.002 mm. that the arc 
drop decreases to about 8 volts and at the same time the neon lines 

Theoretical considerations show that, as a first approximation, 
the light output should be proportional to the current; and actually 
it has been observed by Found that for a large range of values of arc 
current, the light output is 1000 lumens per ampere. At higher 
currents the ratio decreases, and when the discharge is all sodium, 
the light output is only about 500 lumens per ampere. 

In a cathodic type of discharge, the electrons acquire a high kinetic 
energy in the cathode fall region and then lose this energy in excita- 
tion and ionization of atoms. Thus the kinetic energy decreases 
with increase in distance from cathode, and finally when the electrons 
have reached the stage in which their kinetic energy is no longer ade- 
quate for excitation (ultimate electrons) the light generation also 

The fact that the light generation is uniform throughout the 
length of the positive column shows that the electrons must acquire 
energy for excitation and ionization from some other source than the 
cathode fall. This energy is supplied, obviously, by the energy input 
into the column; that is, Gi a watts per unit length, where G is the 



[J. S. M. P. E. 

voltage gradient and i a the arc current. The magnitude of G varies 
with current-density and pressure of gas. Owing to the fact that the 
gradient in a positive column is constant, the concentrations of elec- 



TRA H 3 FORMER, (z) 









FIG. 5. 10,000-lumen a-c. sodium lamp, showing connections for 
operation of cathodes in series with arc. 

trons and positive ions in the column must be approximately equal. 
Since the ions and electrons are constantly diffusing to the walls, 
there must be some mechanism by which fresh ions are generated. 

Jan., 1938] 



Light Output in Positive-Column Neon-Sodium Vapor Discharge. 
In order to obtain considerable light output with appreciable intrinsic 
brilliancy of source, the positive column is more advantageous than the 
cathodic type of discharge. Thus, in the case of low-pressure mer- 
cury vapor and sodium-neon lamps it has been found that from the 
point of view of both lumen output and efficiency the best results are 
obtained with the positive-column type of lamp. 

While the mercury- vapor lamp of this type has been a familiar form 
of illumination for many years, only within the past few years has it 
been found possible to develop a commercial type of sodium-vapor 






Lamp "temperature *C 

tao /so 


FIG. 6 (Left}. Characteristics of 10,000-lumen sodium lamp. 
FIG. 7 (Right). Lumens-watts relation for 10,000-lumen sodium lamp. 

positive-column lamp for use on standard alternating-current cir- 
cuits. Such lamps have been developed by the Osram Company in 
Germany, and by the Philips' Lamp Company in Holland. A lamp 
emitting approximately 10,000 lumens at an input of about 190 watts 
has been described by G. R. Fonda and A. H. Young. 3 Fig. 5 shows 
the design of this lamp and the connections for series-circuit operation. 
The electrodes are oxide-coated tungsten spirals enclosed by elliptically 
shaped anodes, with a distance of about 25 cm. between the latter. 
The pressure of neon used varies from 1 to 3 mm., and under operating 
conditions (6.6 amperes' constant current and a drop of 27-30 volts) 
the pressure of sodium vapor is approximately 0.001 mm. 

The glass is covered internally with a sodium-resistant glaze, and 

72 S. DUSHMAN [J. S. M. P. E. 

the lamp itself is enclosed in a transparent Dewar flask in order to 
maintain the pressure of sodium vapor at the optimum value (which 
corresponds to a temperature in the neighborhood of 230C). 

Now let us consider the phenomena of light production in such a 
lamp. The electrons in a positive-column discharge are distributed 
throughout the whole volume. Therefore, the excitation, ionization, 
and light production are each uniformly distributed along the column. 
At lower pressures of sodium the voltage gradient along the tube is 
higher than at higher pressures, since the ions are supplied for the most 
part by ionization of neon. The sodium is excited by collision with 
electrons that have already lost part of their energy in producing 
the lower-excited states of neon. 

On the other hand, at higher pressures of sodium, the electrons can 
produce a sufficient number of ions by collision with sodium atoms, 
and hence the electron velocities are lower and the voltage gradient is 
decreased. But when this occurs, the wattage input decreases and 
the lamp cools. Hence, for stable operation the lamp must be oper- 
ated at such a pressure of sodium as will maintain the higher voltage 

In this laboratory Found has investigated the operation of the 
lamp as affected by varying the ambient temperature and, conse- 
quently, the vapor-pressure of sodium. In Fig. 6, taken from a re- 
cent paper, 4 the lumens and watts at 6.6 amperes are plotted as 
functions of the external temperature of the tube wall. Below about 
200 C the light is due to excitation of neon. As the temperature is 
increased the light output increases as well as the watts input, and 
both pass through maximum values as shown. 

These observations are plotted also in Fig. 7, which shows the 
light output as a function of watts input. Starting with the cold 
lamp, the lumens increase linearly with the watts, and then remain 
practically constant over a range extending from 207 to about 190 
watts. This characteristic makes it possible, as has been pointed 
out by Found, to operate a neon-sodium vapor-discharge lamp over a 
considerable range of ambient temperature without any appreciable 
change in light output or efficiency. 

The energy distribution for the neon-sodium vapor lamp operating 
at 200 watts' input has been described by Buttolph. 5 Nearly all the 
light occurs in a range between 5600 and 6100 A. It therefore pos- 
sesses an orange-yellow color. 

The average intrinsic brilliancy is about 6 candles per square cm., 


and the efficiency ranges from 50 to 60 lumens per watt, depending 
upon the type of circuit used for operation. 

High-Pressure Mercury-Vapor Lamps. -In the low-pressure gase- 
ous-discharge lamps the average kinetic energy of the electrons is 
very much higher than that of the gas molecules. Thus, while the 
gas in the neon-sodium vapor lamp is at a temperature of about 
230C, the electrons possess a kinetic energy that is the same as that of 
molecules in a gas at about 6000 to 30,000C. This is quite different 
from the state of affairs in the sun or other stars. In the case of the 
latter, thermal equilibrium exists between electrons, ions, and atoms; 
that is, all the constituent particles possess the same average kinetic 
energy. Under these conditions atoms are constantly dissociating 
into ions and electrons while the latter are recombining to form 
atoms, and at constant temperatures the rates of the two reactions 
are equal, so that it is possible to calculate for a given temperature the 
relative numbers per unit volume of undissociated atoms, electrons, 
ions, and excited atoms. 

In an electrical discharge in a gas, conditions approaching those in a 
star are more and more nearly approached as the pressure is increased, 
so that in a discharge in mercury at a pressure of one atmosphere there 
exists a state of approximate thermal equilibrium, and as the pressure 
is increased still higher the temperature in the center of the arc- 
stream gradually increases. 

The earliest form of such a discharge was the quartz tubular lamp 
which has been used mainly as a source of ultraviolet light. By the 
application of thermionic cathodes it has been found possible to de- 
velop designs that are very convenient for practical operation. In 
these lamps a gas, such as neon or argon at a pressure of a few cms. of 
mercury, is used to initiate the discharge between the cathodes which 
are thereby raised to a temperature at which they act as thermionic 
sources. Owing to the high energy input the mercury becomes vapor- 
ized, the pressure increases to one atmosphere or even higher, de- 
pending upon the design of lamp and operating conditions, and the 
spectrum exhibits only those lines that are characteristic of mercury 
along with a certain amount of continuous radiation. 

Measurements of the relative intensities of the lines and of the 
energy distribution in the spectrum made on discharges in quartz tubes 
at different pressures show that while the intensity of the resonance 
line (X = 2537 A) is very high at low currents and low pressures, this 
line is practically eliminated (owing to absorption) in the discharge at 



[J. S. M. P. E. 

pressures of 1 atmosphere and higher. At the higher pressures the 
intensities of the lines X = 3650/3663, X = 5461, andX = 5769/5790 
are increased. In a very recent publication by W. Elenbaas it is 
shown that at higher pressures the spectral lines are broadened more 
and more with increase in pressure, and at higher current-densities 
the continuous spectrum is enhanced. Also, the 
fraction of the total energy radiated in the form of 
near-ultraviolet, visible, and infrared increases with 
increase of pressure, and this accounts for the ob- 
served increase in light efficiency (see subsequent 

The various designs of the high-pressure discharge 
lamp that have been developed may be divided into 
two groups: (1) those operating at 1 atmosphere 
and utilizing glass envelopes, and (2) those operating 
at higher pressures and requiring the use of quartz 

Fig. 8 is a diagram of the 1 -atmosphere lamp 
developed by the General Electric Co., Ltd., Wem- 
bley, England, and described by J. W. Ryde. 6 The 
inner glass tube which carries the arc is enclosed in a 
heat insulating jacket. The wire along the outside 
of the discharge tube is for the purpose of easier 

The electrical characteristics of this lamp have 
been described in a paper by J. A. St. Louis 7 and 
the luminous characteristics by L. J. Buttolph. 5 

The lamp is made in both 400-watt and 250-watt 
units, and operates on a 220-volt a-c. circuit with 
series inductance. Argon is used as starting gas, 
and a limited amount of liquid mercury is inserted 
in the lamp initially. Fig. 9 shows the electrical 
characteristics of the 400-watt unit between the 
time of starting and complete vaporization of the mercury. It will 
be observed that the initial volts and arc watts are low, but that 
after a slight decrease, both the voltage drop in the arc and the 
power input increase rapidly. "During this period of rapidly 
increasing pressure," as St. Louis observes, 7 "the arc constricts. The 
wattage passes through a maximum but the voltage increases with 
time as long as any liquid mercury remains in the tube. After about 

FIG. 8. High- 
intensity 400- 
watt mercury- 
vapor lamp. 

Jan., 19381 



seven minutes all the mercury is vaporized and the electrical charac- 
teristics are thereafter constant. If the mercury were not limited in 
amount, then the course of these characteristic curves would continue 
as indicated by the dotted lines. It should be emphasized that the 
shape of these characteristic curves is influenced by the value of the 
series inductance, by the value of the line voltage, and by the design 
details of the arc tube." 

While it is possible to design the lamp so that equilibrium is attained 
with liquid mercury at a temperature corresponding to any desired 


440 220 
400 ZOO 
360 180 
32O 160 
280 140 


I Z4O 1 20 
5 2OO IOO 
4 |60 80 

3120 CO 
Z 8O 4O 
1 40 ZO 
rt ft n 
































FIG. 9. Electrical characteristics of 400-watt mercury- vapor 
arc between time of starting and complete vaporization of 

mercury-vapor pressure, the use of a limited amount of mercury has 
many practical advantages, not the least of these being the fact that 
under these conditions the lamp is not nearly so sensitive to fluctua- 
tions in either line voltage or ambient temperature. 

Under operating conditions the temperature of the inner glass wall 
is about 350C, the pressure of mercury 1 atmosphere, and the initial 
efficiency 40 lumens per watt. This efficiency decreases after 1500 
hours to about 33 lumens per watt. The average brightness of the 
light-source itself is about 30 candles per square cm. 

The High-Pressure Mercury- Vapor Lamp. One of the most 
interesting of the recent developments in the field of illumination is 



[J. S. M. P. E. 

the quartz capillary lamp developed by C. Bol of the Philips' Lamp 
Co., Eindhoven, Holland. The air-cooled form which operates at 
230 volts with a power input of about 85 watts, consists of a quartz 
capillary 4 mm. in diameter, with a total distance of 18 mm. between 
the oxide-coated tungsten electrodes. A small drop of mercury is 
inserted and argon is used as a starting gas. The lamp is operated 
on a high-reactance transformer having a maximum open-circuit 
voltage of about 450 volts. The pressure of mercury vapor attained 
in operation depends for a given size of capillary upon the gradient 
and the arc current, but is usually greater than 10 atmospheres. 

<o owuu 











After n Hours 










^ \ 






: r 


/.6 1.4 12 I.O 0.8 0.6 0.4 0.2 ( 0.2 O.4 0.6 O.8 I.O 1.2 1-4 1.6 1.8 

mm Radius from Axis 

FIG. 10. Brightness distribution along cross-section of tube. 
Photometric test of water-cooled capillary mercury arc; apparent 
brilliancy of arc viewed through glass jacket. 

By cooling the walls of the quartz capillary in a rapid stream of 
water, the wattage input may be increased more than five-fold, and 
while this increases the efficiency from approximately 40 lumens per 
watt, in the air-cooled, to 60 lumens per watt or even higher, the in- 
trinsic brilliancy and lumen output are increased manifold. In 
fact, Elenbaas 8 has reported that by increasing the power input in the 
water-cooled lamp to 1400 watts per cm. in a 1-mm. diameter tube, a 
brilliancy of 180,000 candles per sq. cm. has been attained. This is 
greater than that of the sun, as observed from the surface of the 
earth, which is 165,000 candles per sq. cm. 

A very comprehensive study of the characteristics of high-pressure 
vapor discharges has been made by Elenbaas and a summary of these 
investigations has been published by G. Heller. 9 


As mentioned previously, the most important difference between 
these discharges in mercury vapor at 1 atmosphere or higher and 
those in much lower pressures is that in the former the temperature of 
the vapor is much higher and thermal equilibrium exists between 
mercury atoms, electrons, ions, and excited atoms, similar to that 
which exists in the atmosphere of a star. 

The temperature in the arc-stream is about 6000 C or even higher 
at the axis and decreases as the walls are approached, so that at the 
latter it is about 800C. As a result, the light distribution across the 
arc-stream is like that shown in Fig. 10 which is based upon some ob- 
servations made in this laboratory by F. Benford and N. T. Gordon. 10 

The pressure of mercury vapor in the lamp increases linearly with 
voltage gradient G for values of the gradient above about 100 volts 
per cm., and at 500 volts per cm. the pressure is about 150 atmos- 
pheres. This condition of operation can be attained only by use of a 
very high rate of cooling. The light output L also increases linearly 
with the watts per cm. of length W, and, as shown by Marden and 
his associates, the relation between the two quantities is of the form 

L = 65 (W - 30) 

Hence, the maximum attainable efficiency is 65 L/W, and for 500 
watts' input the light output is about 30,000 lumens per cm. 

Table III is based upon data published by Marden and his asso- 
ciates in the paper mentioned previously, and presents data on the arc 
characteristics, light output (in candle-power), and average bright- 
ness of the light-source. 


Arc Arc 

Length, Diam., C P. per 

Lamp Volts Amps. Watts C. P. Mm. Mm. Sq. Cm. 

Commercial 150 400 1,560 157 10 100 

High-Intensity Glass 70 250 830 100 8 

Commercial Quartz 250 0.4 85 340 18 1 1,900 

Water-Cooled 580 2.0 920 5,900 18.5 1.5 21,000 

Water-Cooled 840 2.08 1490 10,500 17.5 0.85 70,000 

As shown by results obtained in this laboratory by F. Benford and 
N. T. Gordon, it would seem quite practicable to operate a water- 
cooled quartz lamp at 500 volts per cm. and 1 ampere, with an 
average brightness of 20,000 candles per sq. cm. 

78 S. DUSHMAN [J. S. M. P. E. 

The color of the light emitted has considerably more red than the 
low-pressure lamp and thus shows a better color reproduction. 

It should be mentioned in this connection that in the case of the 
water-cooled lamp operated by Elenbaas which gave a brightness of 
180,000 candles per sq. cm., the power input was 1400 watts per cm. 
in a tube 1 mm. in diameter; the gradient was 805 volts per cm. ; and 
the mercury vapor pressure attained was 200 atmospheres ; while the 
temperature at the axis was calculated as 8600C. 

Transformation of Ultraviolet Radiation into Visible Light. That 
many materials fluoresce under the action of ultraviolet radiation is a 
fact that has been known for a long time, and quite a vast literature 
has accumulated describing both the methods of preparation of vari- 
ous "phosphors" and the relation between the emission spectra and 
the wavelength of exciting radiation. Some materials, like the sul- 
fides of the elements of Group II in the periodic table (Lenard 
phosphors), respond to radiation in the range X = 3000 A to X = 
4000 A, while others, such as the silicates of zinc and cadmium and 
some of the tungstates and molybdates, respond better to radiation 
in the neighborhood of X = 2537 A (the resonance line of mercury). 
In the case of the sulfides as well as the silicates, the fluorescence is 
observed only if some "activator," such as bismuth or copper in the 
first case or manganese in the latter, is present. A list of the more 
commonly used phosphors, as compiled by Fonda, of this laboratory, 
is given in Table IV. 


Exciting Radiation Emitted Fluorescence 
Phosphor Activator Range Peak Range Peak 

Zinc Silicate Manganese 2200-3000 2530 4600-6000 5100 
Cadmium Silicate Manganese 2200-3200 2530 5200-6500 5900 
Calcium Tungstate Lead 2200-3000 2500-2800 4300-5150 5200 
Magnesium Tung- 
state Lead 2200-3300 2500-3000 4300-6500 5400 
Zinc Sulfide Copper 2400-4400 3600-4300 4700-6200 5400 
Zinc Cadmium 

Sulfide Copper 2400-4400 3600-4300 5100-6700 5800-5900 

The second column gives the activator, while the other columns give 
the wavelengths of exciting and emitted radiations. 

Fig. 11 shows observations made by Fonda on the excitation and 
emission spectrum of the zinc silicate. It will be observed that the 
two spectra are separated by quite a wide interval, which corresponds 

Jan., 1938] 



to about 1.4 electron volts. On the other hand, in the case of Lenard 
phosphors the maximum wavelength for excitation is adjacent to the 
minimum wavelength in the emission spectrum. 

Furthermore, a large number of organic compounds, such as eosin, 
fluorescein, and rhodamine B, have been found to exhibit fluorescence. 
The last-named compound emits radiation in the orange-red under the 
influence of radiations in the near-ultraviolet region, and has there- 
fore been suggested for use with the quartz capillary lamp described 
previously. In this case the dye is painted on a reflector surrounding 
the lamp. 

During recent years the problem of utilizing fluorescent materials 
to increase the efficiency of light-sources has received considerable 
attention. In Germany, M. Pirani and his associates in the labora- 


5000 3200 46OO SOOO MOO S4OO SMO MOO 6OOO 62OO 69OO 

FIG. 11. Excitation and emission ranges of wavelength for 
activated zinc silicate. 

tories of the Osram Company have made notable contributions in 
this field, and in this country G. Inman of the Incandescent Lamp 
Division of the General Electric Co., in Cleveland, has recently de- 
scribed a type of fluorescent lamp that is a highly efficient source of 

A positive-column discharge is passed through mercury vapor at 
low pressure (corresponding to the vapor pressure at slightly higher 
than room temperature). This constitutes a very efficient source of 
the mercury resonance radiation, and, by coating the inside of the 
discharge tube with a silicate or tungstate, light in the visible range 
is obtained. The color emitted and the luminous efficiency vary with 
the composition and mode of preparation of the fluorescent material. 
Thus, by using a specially prepared zinc silicate, it is possible, accord- 
ing to Inman, to obtain light of a green color at an efficiency of 60 
lumens per watt and even higher. 


Many of these phosphors, as, for instance, the silicates and sul- 
fides, continue to emit light for a short interval after the exciting 
source has been removed. This phenomenon is known as phos- 
phorescence, and phosphors that exhibit this effect can be operated on 
a 60-cycle circuit without any observable flicker. 

From the point of view of an interpretation of the mechanism of 
fluorescence these observations should prove important. While our 
understanding of these phenomena is very indefinite at the present 
time, there is no doubt that the whole problem is intimately related to 
the presence of energy bands and localized levels in crystalline solids. 
Consequently, we may expect that further investigations will be of 
material assistance in interpreting the phenomena of fluorescence and 
phosphorescence in general. 


1 DUSHMAN, S.: "The Search for High-Efficiency Light-Sources," /. Opt. 
Soc. Amer., 27 (Jan., 1937), No. 1, p. 1. 

2 FOUND, C. G.: "Fundamental Phenomena in Sodium-Vapor Lamps," Gen. 
Elect. Rev., 37 (June, 1934), No. 6, p. 269. 

3 FONDA, G. R., AND YOUNG, A. H.: "The A.C. Sodium-Vapor Lamp," Gen. 
Elect. Rev., 37 (July, 1934), No. 7, p. 331. 

4 FOUND, C. G.: "Fundamentals of Electric Discharge Lamps," Trans. Ilium. 
Eng. Soc. (Presented at White Sulphur Springs, W. Va., September 29, 1937). 

5 BUTTOLPH, L. J. : "High-Intensity Mercury and Sodium Arc Lamps," /. Soc. 
Mot. Pict. Eng., XXIV (Feb., 1935), No. 2, p. 110. 

6 RYDE, J. W.: "The Electrical Characteristics of the New 'Osira' Lamp," 
Gen. Elect. Co. J. (England), 4 (Nov., 1933), p. 199. 

7 ST. Louis, J. A.: "Characteristics of 400- Watt and 250- Watt Type H Mer- 
cury Lamps," Trans. Ilium. Eng. Soc., 31 (June, 1936), No. 6, p. 583. 

8 ELENBAAS, W. : "liber die mit den wassergekuhlten Quecksilber-Super- 
Hochdruckrohren erreichbare Leuchtdichte," Zeit. filr techn. Physik, 17 (Feb., 
1935), No. 2, p. 61. 

9 HELLER, G. : "Dynamical Similarity Laws of the Mercury High-Pressure 
Discharge," Physics, 6 (Dec., 1935), No. 12, p. 389. 

10 A similar curve is shown by MARDEN, J. W., BESSE, N. C., AND MEISTER, G.: 
"Brightness of the Mercury Arc," Ilium, Eng. Soc. (Presented at White Sulphur 
Springs, W. Va., September 27, 1937). 






Summary. This report presents therevised specifications of the Research Council's 
standard electrical characteristics for two-way reproducing systems in theaters, as 
well as other Research Council standards relating to power reference level, cross-over 
frequency, output requirements for theaters, and standardization of harmonic content. 

In addition, data on the Research Council standard frequency test-film and theater 
test-film are presented, as well as specifications for the proposed standard nomen- 
clature for filters. 


Since the addition of recorded sound to motion pictures, the major 
studio sound directors have recognized a need for standardization of 
theater sound projection equipment in order that there would be a 
practical uniformity of product from all companies regardless of the 
theater in which it was reproduced. Uniformity and standardization 
of sound projection equipment would obviously react beneficially to 
the entire industry and would make certain that the character and 
expression put into the film would be reproduced in all theaters meet- 
ing the standard conditions. 

Recognizing this, the Research Council of the Academy of Motion 
Picture Arts and Sciences, upon the recommendation of the sound 
directors, appointed a Committee to undertake a study of the problem 
of theater sound equipment standardization. 

It was realized that one of the first objectives of this project should 
be the establishment of a standard electrical characteristic to which 

* Presented at the Spring, 1937, Meeting at Hollywood, Calif., during a meeting 
sponsored by the Academy Research Council, May 27, 1937, at Metro-Goldwyn- 
Mayer Studios; received Aug. 13, 1937. 




[J. S. M. P. E. 

the equipments of all theaters might be set. After a series of tests 
and necessary compromises, the Council, upon the recommendation 
of this Committee, adopted the Standard Electrical Characteristic 
for two-way reproducing systems in theaters described below as be- 
ing the most suitable for -present conditions. 

It has been customary in the past to adjust theater reproducing 
equipment to satisfy the ears of individual groups. To obtain these 
results, test tracks of various characteristics made by the separate and 
many organizations were used. Since the adjustments made with a 
sound-track from one organization did not always meet with the ap- 
proval of other organizations, they in turn modified their character- 

FIG. 1. Research Council standard electrical characteristic for two-way 
reproducing systems in theaters (June 8, 1937). Electrical run measured 
at the output of the power amplifier with a response equivalent to the speaker 
load, using ERPI test-film (ED-20, corrected), or RCA test-film (cat. No. 

(Curve applying to non-metallic (bakelite) diaphragm supersedes curve in 
specifications of March 31, 1937. Curve applying to metal diaphragm re- 
mains unchanged.) 

istic to obtain optimal results from this theater adjustment, which, as 
a consequence, was ever changing. This practice created a vicious 
cycle of theater adjustment and studio compensation adjustment. 
It was finally recognized by those familiar with and responsible for 
the projection of studio sound recording that this situation was be- 
coming more and more impracticable. 

In inaugurating its program, the Committee prepared a test-reel 
containing a 250-ft. section of release print from each studio, so chosen 
that the assembled reel contained representative examples of both 
dialog and music recordings made under average as well as under 
extreme conditions by each studio sound department. 


A series of test runnings was made with this reel at the Carthay 
Circle Theater in Los Angles, Grauman's Chinese, the Filmarte, 
Oriental, Pantages, and Warner Brothers' Hollywood Theaters in 
Hollywood, during which the electrical characteristics of the equip- 
ment installed in each of the theaters were varied in order that the 
Committee might determine an optimal electrical characteristic that 
would most nearly fit the acoustic characteristics of this group of 
theaters. These particular theaters were chosen as having divergent 
characteristics to which a standard might be fitted, the Committee 
operating upon a premise that a standard that would fit these would, 
in general, fit at least a majority of the theaters throughout the coun- 

Fig. 1 shows the electrical characteristics for theaters that have 
been adopted as standard by the Academy, having been approved 
by the sound directors of all the major studios, the sound equipment 
companies, and the Academy Research Council. In the opinion of 
the Committee, this characteristic will give the best reproduction of 
the film product from all studios today. It covers a frequency range of 
50 to 8000 cps. with reductions in volume at the upper and lower ends 
of the range. These are necessary to minimize the effects of noise 
and extraneous signal material introduced mostly in the later links 
of the recording processes. 

As improvements are made and the recording characteristics 
changed in the studio, similar compensation can be made in all modern 
theater reproducing systems at little or no cost for additional equip- 

Some of the factors that made it desirable to depart from linearity 
for the standard electrical characteristics follow: 

(1) Film and system noise of high-frequency content. 

(2) High-frequency extraneous noise caused by phase shifts and intermodu- 
lated effects. The magnitude of these depends upon the type of light-modulator 

(5) Variable dynamic high-frequency distortion effects caused by the variation 
in average slit-width due to both signal and noise-reduction components during 

(4) Flutter due to improper motion of the film. (This limitation is rapidly 
being improved with the current change to new sound -heads.) 

(5) Extremely low-frequency components introduced by the noise-reduction 

In the early days of the application of noise-reduction systems, 



[J. S. M. P. E. 

efforts were made to improve the reproduced signal-to-noise ratio in 
the upper part of the frequency range by pre-equalization and post- 
equalization methods. 

As is well known, this consists in the use of equalizers to increase 
the relative upper-frequency level before the signal is recorded, and 
the employment of other post-equalizers in the reproducing chain to 
restore the signal to normal. The results attained, however, were not 
a definite improvement, because for light-valve operation in the high- 
frequency range the reproduced volume does not vary in a linear 
manner with the signal volume applied to the light-modulator. Post- 

FIG. 2. Loss of high-frequency volume level with increasing modulation, 
recorded with four-ribbon valve, one-mil spacing, no back stop rectifier in 
noise-reduction amplifier. 
(1) Full modulation 

6 db. from full modulation 
12 db. from full modulation 
21 db. from full modulation 
30 db. from full modulation. 


equalization thus became impracticable because there were required 
changes in equalization with changes in volume. For the galvanom- 
eter type of modulator with either variable-width or variable- 
density type of recording these non-linear restrictions do not apply, 
and it may be possible in the future to use pre- and post-equalization 
with this type of equipment. 

Fig. 2 shows the response of a light-valve from full modulation to 
30 db. This test was recorded on a four-ribbon push-pull light- valve 
having a 1-mil fixed spacing, and then biased to a 0.2-mil spacing by 
the noise-reduction amplifier. The biasing circuit was so arranged 
that any signal requiring more than 100 per cent modulation would 
drive the ribbons beyond the normal 1-mil spacing. 

This practice causes a wide variation in slit width, which in turn 
causes a decrease in high-frequency output. When the effect of slit 


width variation is reduced by decreased ribbon movement, or its 
equivalent optically, this change in response is decreased at some ex- 
pense in output level. 

The specifications for the standard electrical characteristic dis- 
cussed above follow: 


(Revised specifications superseding specifications of March 31, 1937) 

Systems to Which These Specifications Apply. The two-way repro- 
ducing systems for which this characteristic, indicated below and by 
the associated curve which is a part of these specifications, is recom- 
mended, are: 

Type I. Mirrophonic system using 594- A mechanisms (loud speaking tele- 
phones) (metal diaphragm) and TA-4181-A low-frequency mechanisms (loud 
speaking telephones). 

Type II. RCA system using MI-1435 (metal diaphragm) and MI-1432-A low- 
frequency mechanisms. 

Type III. RCA Lansing equipped system using 284 (metal diaphragm) and 
15X low-frequency mechanisms. 

Type IV. RCA system using MI-1428-B (non-metallic diaphragm) and 
MI-1432-A low-frequency mechanisms. 

Measurement Point. This characteristic is valid for measurements 
made at the output of the power amplifier, including the low-pass 
filter, with a resistance equivalent to the speaker load, using the 
Electrical Research Products, Inc., test-film ED-20 (corrected),** or 
the RCA test-film Catalogue No. 27637, and is subject to modifica- 
tions to fit special acoustic conditions that no doubt exist in many 
theaters, due to the fact that the reverberation time or other acoustic 
characteristics are not optimal. 

Gain- Frequency Characteristic. The following table indicates the 
characteristic for both the metallic and non-metallic types of dia- 
phragms used on the high-frequency mechanisms : 

* Reprinted from the Bulletin of the Research Council, Academy of Motion 
Picture Arts and Sciences, June 8, 1937. 

** The correction factor, printed on the back of the can in which this test-film 
is furnished, indicates the deviation from constant percentage modulation for 
each frequency. 

J. K. HlLLIARD [J. S. M. P. E. 

Frequency Metal Diaphragm Non-Metallic Diaphragm 

Mechanisms Mechanisms 

50 -1 to - 3* -1 to -3 

100 - V* to - 1** - V* to -1 




2000 /4 + Vi 

3000 I 1 /* +1 

5000 4 1 /* +1 

7000 lO 1 /* -2 1 /* 

8000 18 -6 

Tolerance. A tolerance of =*= 1 db. is specified for any of the above 
gain-frequency measurements. 

Acoustic Correction. Whenever such conditions exist that this 
characteristic does not give satisfactory results, it is recommended 
that the acoustic characteristics of the auditorium be corrected. 

Mechanism Adjustment. With the presently available equipment 
as specified, operating with the standard electrical characteristic, it 
is necessary in some instances that the sensitivity of the high- and 
low-frequency bands be relatively adjusted to obtain a flat acoustic 
response on both sides of the cross-over. This adjustment usually 
takes the form of attenuating the high-frequency band by means of 
the taps in the dividing network to varying degrees from to 5 db., 
depending upon the relative efficiency of both low- and high-frequency 
units and the specific acoustic properties of the auditorium involved. 
Typical values are as follows : 

ERPI, Mirrophonic system: attenuate the high-frequency band 
2 to 4 db. 

RCA, MI-1435 and MI-1432-A: attenuate the high-frequency 
band to 2 db. 

RCA, Lansing equipped : attenuate the high-frequency band to 

RCA, MI-1428-B, MI-1432-A: attenuate the low-frequency band 
to 2 db. 

Note. It should be remembered that the type and condition of 
screen used in the theater will in a measure affect the high-frequency 
response of the reproducing system. 

* For M-3 and M-4 Mirrophonic Systems, 50 Cycles, + 1 to 1. 
** For M-3 and M-4 Mirrophonic Systems, 100 Cycles, + V* to */* 




During the past few years there have been several standards for 
reference power level, such as 6, 10, and 12 milliwatts, which fact has 
led to certain confusion in the interchange of knowledge of equipment. 

It has been considered desirable to adopt a single reference stand- 
ard, which by common agreement has been designated 6 milliwatts, 
and all organizations have agreed to rate and measure all equipment 
in terms of this reference. 


The distortion present in a horn is directly proportional to its 
length, and in the early development of the two-way speaker, effort 
was made to construct the high-frequency horn as short as possible. 


FIG. 3. Horn frequency characteristics. 

The theoretical cut-off was taken as approximately 200 cps., and net- 
works were used with a cross-over as low as 250 cps. 

However, after continued tests certain distortion was noted, and 
subsequent tests indicated that a higher cross-over would be desirable, 
300 cps. in some, and as high as 400 cps. in other, equipment. Re- 
cently further tests have been completed and the results have indi- 
cated that a cross-over no lower than 400 cps. will give optimal perform- 
ance. This is undoubtedly due to the uniform impedance presented 
by the horn at frequencies above this point, which does not apply to 
frequencies within an octave of the theoretical cut-off. Fig. 3 shows 
the change of impedance of a horn near the cut-off frequency. 


J. K. MILLIARD [j. s. M. P. E. 


The use of the wide volume range film that is now being released 
requires that the theater reproducer have sufficient output capacity 
and efficiency to reproduce this volume range adequately and with- 
out compression. 

The history of the reproduction of sound has been one of continual 
increase in amplifier carrying capacity. Originally output powers 
from 2.5 to 12 watts were considered adequate for the volume range 
encountered. Since the studios have found that it is necessary to have 

FIG. 4. Amplifier capacity assuming output of one acoustic watt per 1000 
sq. ft. of floor area (RCA). (Note. The auditorium must be adjusted for 
optimal reverberation time.) (Two electric watts equal one acoustic watt.) 

at least a 60-db. volume range for future requirements, it has been con- 
sidered necessary to increase the power-carrying capacity by large 
amounts. Sound-effects involving screams, earthquake noises, gun 
shots, and other sounds incident to warfare demand sensation levels 
considerably higher than those that could be delivered in the past. 
For that reason, a maximum output level of not less than 90 sensation 
units is now considered necessary (whereas in the past, amplifier 
carrying-capacity had been limited to 80 db. above the threshold of 
hearing) . 

Accordingly, to obtain a yard-stick to measure the power necessary 
for a theater when either the floor area or the cubical content is known, 
Figs. 4 and 5 indicate the installed amplifier capacity necessary to 



maintain the standard required. Since the required power is a func- 
tion of the absorption or reverberation in the theater, the curves are 
based upon a condition of optimal reverberation. In practice, there- 
fore, deviation will be required, depending upon the variation from 
the optimum. 

The optimal reverberation time at 512 cps. is shown for auditoriums 
of various volumes in Fig. 6. Fig. 7 shows the optimal reverberation 
times in the frequency range used in recording, for an auditorium of 
approximately 300,000 cubic feet capacity. 


With the increase of fidelity of theater reproducing systems it has 
been found necessary to standardize the load-carrying capacity of an 

FIG. 5. Installed recommended electric watts/cubical content (ERPI). 

amplifier. The standardized rating has been taken at that point at 
which the amplifier introduces 1 per cent third harmonic or 2 per cent 
total harmonics. 


The application of the standard electrical characteristic for two- 
way theater reproducing systems is measured in terms of a standard 
frequency test-film. 

At the time this standardization program was started there were 
several test-films prepared by various organizations, the use of which 
required a correlation of their calibration. Before the Committee had 
gone very far it appeared that the adoption of a new single test-film 
would be advantageous to all concerned. 



fj. S. M. P. E. 

A variable-width test track was recorded with one of the latest 
variable-width recorders using ultraviolet light for recording and 
printing. This negative was then circulated among several labora- 
tories, and all the prints were measured on a common reproducing 
channel with known characteristics as shown. 

Fig. 8 shows the variation in level for the different frequencies as 
indicated on a continuous level-recorder (it will be noted that the 
variation is less than 0.2 db. within any one frequency). Fig. 9 shows 
the film output from a standard RA-1010 ERPI recorder, and Fig. 10 
shows the response from different laboratories. 





g 1.0 






J ^ 






i ^* 


_ --- 

P 9 c 

OO f\J |k C 

00 10,000 100.000 I.OOC 


FIG. 6. Optimal reverberation time vs. volume in cu. ft. for 512 cps. (after 


A standard warble frequency test-film is also available using 24 test 
frequencies from 40 to 10,000 cps., having a 5 per cent warble. This 
film is used to determine the acoustic output of the speaker system. 
In addition, a warble continuous-sweep test-film from 40 to 10,000 cps. 
is available. 

When this variable- width frequency film (known as Test No. 1775) 
is used, the following corrections should be applied to the readings 

Cycles Db. Cycles Db. 

440 +0.4 1,000 -0.0 

80 +0.4 3,000 -0.6 

150 +0.2 5,000 -2.4 

300 +0.3 7,000 -6.6 

375 +0.3 8,000 -6.0 

500 +0.1 10,000 -82 



The plus sign indicates modulation above normal and the negative 
sign indicates modulation lower than the reference point. With these 
corrections applied the film becomes equivalent to the ERPI ED-20 
(corrected) or the RCA test-film No. 27637. 


The original test-film comprising sections of release prints from each 
studio has been used by the Committee for a period of six months, 
and changes in recording technic required to fit the standard electrical 

FIG. 7. Optimal reverberation times in the frequency range used in re- 
cording, for an auditorium (Carthay Circle Theater in Los Angeles) of ap- 
proximately 300,000 cu. ft. 

characteristic have made this particular test-film obsolete and a new 
one has taken its place. 

The Research Council will soon make available to the industry a 
Movietone test-film containing samples of dialog and music recording 
from each of the eight major studios, so chosen that the assembled 
reel will contain representative examples of sound recorded under 
average as well as under extreme conditions by each studio sound 

This film will be similar in make-up to the test-film used by the 
Committee in arriving at the standard electrical characteristic, and 
will be extremely useful in the field for routine checking and main- 
tenance of adjustment of the theater sound systems. 

Prints of the film may be obtained for a nominal fee by sound 



[J. S. M. P. E. 

equipment service companies, theater circuits, and other organiza- 
tions concerned with the maintenance of sound quality in the theater.* 


In consideration of the confusion arising from the variety of 
methods by which wave-filters are designated in the field, this Com- 
mittee recently undertook, as a second step in its program, the stand- 
ardization of filter nomenclature. 

The following quotation from the report of the Committee** pre- 
pared for consideration of the Research Council will best describe this 
portion of the work : 

"At the present time there are two general methods for designating 
filters, neither of which conveys such information as is needed to 


FIG. 8. Variation in level with frequency, of standard fre- 
quency test-film, as indicated on continuous level recorder. 

establish the filter characteristics, and it was consequently recognized 
by the Committee that in addition to adopting a standard, any 
method worked out should convey definite information concerning 
the limits of the transmission bands. 

Since the presentation of this paper, the proposed standard nomen- 
clature for filters has been approved for use in the theater field by 

* Since the presentation of this paper, this reel has been completed, and prints 
are now available. Inquiries should be addressed to the Research Council of 
the Academy of Motion Picture Arts and Sciences, Suite 1217, Taft Building, 
Hollywood, Calif. 

** Bulletin of the Research Council of the Academy of Motion Picture Arts and 
Sciences, August 10, 1937. 





_ - 










TIC ( 

r R 

*, K 



E R 

f 1. RERECO 









s s 











I 1000 


FIG. 9. Reproduced output of standard test-film from RA-1010 ERPI re- 

FIG. 10. Response curves of prints of standard test-film made at four 
laboratories. (Note. The print from laboratory A (indicated as flat from 
40 to 10,000 cps.) has been arbitrarily chosen as the print to which the others 
are referred. The chart shows deviations of the other prints from print A.) 



[J. S. M. P. E. 




















? ? 

















? T 





















Electrical Research Products, Inc., the RCA Manufacturing Com- 
pany', and technical representatives of many of the theater companies 
cooperating in this standardization program, and for use in sound 
recording circuits by the sound directors of all the major producing 
companies; and has subsequently been approved by the Research 
Council of the Academy of Motion Picture Arts and Sciences as an 
industry standard effective August 15, 1937. 

"Both methods now in use for designating filters employ the fre- 
quency that separates the transmission range from the suppression 
range. For band filters two such separation frequencies are necessary, 
while for low-pass and high-pass filters only one is needed. Inasmuch 
as the insertion loss of a filter changes gradually in the cross-over 
region, the specification of a separation frequency is a matter of defini- 
tion. The two methods now used differ from each other in their man- 
ner of defining these frequencies one method defines the separation 
point as the frequency at which a 10-db. insertion loss is obtained, 
whereas the other method uses theoretical cut-off frequencies. 

"Neither of these methods conveys sufficient information regarding 
the insertion loss characteristic of filters within their transmission 
band. The 10-db. loss method does not give information as to the 
manner in which the insertion loss characteristic approaches this 
point; and the theoretical cut-off frequency method gives no loss in- 
formation whatsoever, although anyone familiar with the design of 
filters can visualize roughly the manner in which the change occurs. 

"Consequently this Committee recommends that both the meth- 
ods described above be discarded and that a standard nomenclature 
for filters, as specified, be used exclusively hereafter : 

"Specification. The standard symbol describing any filter shall 
consist of three characters, the first designating the frequency of 3-db. 
insertion loss ; the second the character Hi or Lo to indicate high-pass 
or low-pass; and the third the frequency at 10-db. insertion loss (all 
frequencies in cycles). 

"Thus, the following describes several low-pass filters: 4000 Lo 
6000 (Fig. 11); 5000 Lo 7000 or 4500 Lo 5500 and the following describe 
several high-pass filters: 60 Hi 40 (Fig. 12), 80 Hi 30 or 100 Hi 50. 

"It might be pointed out that a combination of two of the above 
symbols may be used to describe a band-pass filter (Fig. 13) or a divid- 
ing network (Fig. 14), or a reverse combination of symbols may be 
used to describe a band-elimination filter (Fig. 15)." 



Summary. A new sound-on-disk recorder has been developed in which is used 
the principle of feeding part of the output of the system back to the input of the associ- 
ated driving amplifier in properly controlled relationship. The use of this principle, 
which is widely used in feedback amplifiers, replaces the usual practice of providing 
dissipative elements for the control of an electrically driven vibrating system. Here- 
tofore no practical application of feedback to electromechanical systems has been made, 
possibly because the requirements for stable operation oj such systems are difficult of 
achievement. Through recent developments these requirements have been satisfac- 
torily met. The new recorder is capable of recording on wax or direct-recording ma- 
terial without appreciable effect upon its characteristics, which include uniform re- 
sponse from 30 to 12,000 cps. and exceptional freedom from distortion. The recorder 
is extremely simple and affords easy means for field calibration from the feedback 
element, whose output is in direct proportion to the stylus velocity. These means also 
make available a monitoring voltage which, properly amplified, gives a precise aural 
picture of the stylus behavior during recording. 

The successful application of new principles to a design for an elec- 
tromechanical recorder to be discussed here offers great possibilities 
in the general treatment of electromechanical transducer problems. 
In the usual treatment of these problems, resonant and dissipative 
elements, each carefully controlled within predetermined limits, com- 
prise the vibratory system. The high degree of perfection to which 
this procedure has been followed in certain types of disk recorder 
may be judged by the data given previously by H. A. Frederick. 1 
The simplicity of the new device, whose individual elements require 
control only within broad limits, is made possible through the prin- 
ciple of regenerative feedback, an added advantage of which is the re- 
duction of distortion and noise components arising in the recorder and 

Broadly speaking, feedback may be defined as coupling from the 
output of an amplifying system to its input. During the early devel- 
opment of high-gain amplifiers, the avoidance of feedback constituted 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received October 
1, 1937. 

** Bell Telephone Laboratories, New York, N. Y. 



a serious problem. The familiar singing or howling was a common 
and discouraging manifestation of uncontrolled feedback, which was 
ultimately eliminated by improved shielding and wiring methods. 
Feedback, properly controlled, found very early uses in oscillating 
and other forms of regenerative electrical systems. In January, 
1932, H. Nyquist 2 published the conditions necessary for stabilizing 
regenerative circuits and in January, 1934, H. C. Black 3 discussed 
practical feedback amplifiers in which these conditions were met. 

Electromechanical Feedback System. The theory of a feedback sys- 
tem for an electromechanical device is not unlike that for an amplifier 
except that the relations must include factors for the electromechani- 
cal conversion of energy. Fig. 1 shows diagrammatically a feedback 
disk recorder-amplifier system. The purpose of the system is to 






E 3 

- \ 

FIG. 1. Schematic representation of an electro- 
mechanical feedback system. 

move a cutting stylus in a recording medium with a vibrational ve- 
locity whose wave-shape is an exact replica of the wave-shape of the 
signal voltage. The output voltage E 2 of the amplifier is supplied to 
the terminals of the recorder, thereby driving the stylus with a 
velocity V. The motion of the stylus in turn generates the voltage 
3 by means of a suitable generating element such as a small coil mov- 
ing in a magnetic field. This voltage is returned to the amplifier in- 
put through a control circuit which may be either passive or active. 
The voltage available after modification in the control circuit is desig- 
nated 4 , and adds to the signal voltage E. It must be mentioned 
that the voltages and velocities here referred to are to be considered 
as having both magnitude and phase, and hence can be represented 
in complex number notation. The sum of E and E^ which is the 
voltage EI actually applied to the amplifier, may be greater or less 
than the signal voltage E depending upon the phase relation of E and 

98 L. VlETH AND C. F. WlEBUSCH [J. S. M. P. E. 

4 . If the sum is less than the signal voltage alone, the system is 
said to have negative* feedback; if the sum is greater, it is said to have 
positive feedback. 

To obtain a simple expression for the relation of the stylus velocity 
V to the signal voltage E, let 

and -- 

and hence AB = E 4 /i (3) 

The product AB thus defines the transmission around the loop 
formed by the amplifier, recorder, and feedback control. The value 
of 4 from this equation can now be substituted in the relation EI = 
E + E 4 to obtain 

1 A* 

1 AB 
which, together with equation 1, gives 

The factors A and B can be calculated from a knowledge of the 
elements of the amplifier, the feedback control, and the recorder; or 
they may be individually measured for an existing structure. Equa- 
tion 5 permits the calculation of the overall performance of the system ; 
that is, its amplitude vs. frequency characteristic and its phase-shift 
vs. frequency characteristic. Several interesting and instructive con- 
clusions may be drawn from the relations just discussed. It is ob- 
vious that if, at any frequency, the quantity AB becomes equal to 1 
+ jO, the denominator of equation 5 becomes zero and the system will 
sing or oscillate. Actually the condition for stability is somewhat 
more complicated than the mere avoidance of an AB product of ex- 
actly unity. Nyquist showed that for stable operation of a system 
such as is here considered, a polar plot of AB \ |_0_ and its conjugate 
from zero to infinite frequency must not enclose the point 1 | . Fig. 
2 (a) is a plot of this factor for a typical stable system, and Fig. 2(6) 
is for a possible unstable one. It is apparent that to minimize the 
danger of singing, the phase-shift must be kept well within the limits 

Jan., 1938] 



of 180 =*= 180 degrees for all frequencies for which there is a transmis- 
sion gain around the loop, that is \AB\ > 1. 

Of greatest interest, perhaps, is the effect of feedback upon the 
frequency characteristic. If the feedback is zero, B is zero, and the 
system will perform as a simple amplifier and recorder. If B is now 
increased until, over some chosen frequency range, the magnitude of 
AB is large compared to unity, equation 5 becomes 


which indicates that over the frequency range considered the velocity 
of the stylus is independent of the amplifier gain or the efficiency of 
the recorder. Variations in B, however, directly affect the perform- 



FIG. 2. Typical polar plots of the factor | AB \ (^ (a) For a stable 
feedback system. (6) For an unstable feedback system. 

ance, and hence if a flat frequency response is desired, B must remain 

constant. However, since B is the product of the mechanical-electri- 
cal conversion factor E S /V and the control factor E^/E 3 it will be 
seen that these factors may vary so long as their product remains con- 
stant. As indicated later it is a simple matter to maintain the factor 
JE 3 / V constant and hence a flat response characteristic depends only 
upon keeping the control factor constant. 

If equation 5 is rewritten to include noise and distortion products 
as well as signal, it becomes 

I i 

I - AB I - AB I - AB 


100 L. VlETH AND C. F. WlEBUSCH [J. S. M. P. E. 

where n and d are the noise and distortion, respectively, introduced in 
the amplifier and recorder without feedback. Hence, when AB is 
large compared to unity, both the noise and the distortion components 
are reduced as compared with the corresponding effects in a non-feed- 
back system. 

Forces acting upon the stylus during cutting may be regarded as 
noise or distortion introduced in the recorder and their effect upon the 
vibrational velocity is also reduced by the above factor. This is 
equivalent to a manifold increase in the stiffness and massiveness of 
the moving element. 

Requirements for an Electromechanical Feedback System. From what 
has been said it will be apparent that the requirements to be met by 
the feedback amplifier-recorder system in order to realize the fore- 
going results are : 

(1) The voltage wave from the feedback generating element, actuated by the 
stylus velocity, must be the exact replica of the stylus velocity wave. 

(2) The sum of the phase-shift contributed by the electrical to mechanical to 
electrical conversion in the recorder, and the phase-shift contributed by the 
amplifier and that contributed by the feedback control circuit, must be well within 
the limits for stable operation. 

(5) The power capacity of the amplifier and of the recorder must at any fre- 
quency be sufficient to drive the stylus at the desired velocity. 

The third requirement, of course, applies whether or not feedback 
is involved, and it is desirable from a power economy standpoint to 
make the electromechanical conversion efficiency as high as possible 
within the desired frequency range. 

Description of Electromechanical Transducer or Recorder. The first 
requirement may be met by assuring a rigid connection between the 
stylus whose velocity is to be controlled and the feedback generating 
element. A moving-coil generating element and a moving-coil drive 
suggest themselves as a convenient way of meeting the second or 
phase-shift requirement. The principal source of phase-shift in 
such an electromechanical transducer is the mechanical vibrating 
system itself; that is, the phase relation between driving force and 
resultant velocity. A singly resonant system offers the most advan- 
tageous solution because its inherent phase-shift for frequencies from 
zero to infinity is only 90 degrees. A more complex system ob- 
viously would be less desirable from this standpoint. The fulfill- 
ment of the third requirement involves well-known principles and 
will not be discussed. 

Jan., 1938] 



Fig. 3 is a cross-sectional view of a portion of a recorder that em- 
bodies all these features and meets all three requirements. The driv- 
ing coil is secured at the base of a cone-shaped vibrating element, the 
restoring force being furnished by a cantilever spring and a diaphragm, 
which serve also to restrict motion of the stylus to one mode. A sec- 
ond coil, the feedback coil, is secured to the cone near its apex, at 
which point the stylus is attached. The coils are free to move in an- 
nular air-gaps polarized by a common magnet. In the space between 
the two coils copper shielding is provided to reduce magnetic coupling. 
The output of the amplifier is supplied to the driving coil and the out- 

FIG . 3 . Cross-sectional view of vibrating system and closely 
associated magnetic circuit of the feedback recorder. 

put of the feedback coil is connected through the control circuit to 
the input of the amplifier. 

In the actual design the device is provided with the usual acces- 
sories for disk recording, such as suction pipe, advance ball, etc. A 
photograph of the complete recorder is shown in Fig. 4. 

With the recorder properly wired into the system the technic of re- 
cording is exactly the same as with non-feedback recording systems. 
Inasmuch as the feedback system provides a flat response, any de- 
sired alteration of the response characteristic may be accomplished 
by means of electrical equalizers either ahead of the feedback system 
or in the feedback control circuit. 

As shown in Fig. 4 connections to the recorder are made through 
concentric jacks which terminate the leads from the driving coil and 



[J. S. M. P. E. 

the feedback coil. Concentric wiring is used to avoid coupling be- 
tween the circuits. 

Performance Characteristics. Curved, Fig. 5, shows the frequency 
response characteristic of a laboratory model of the feedback ampli- 
fier-recorder system with the feedback circuit opened; i.e., B = 0. 
Curve B shows the same characteristic with proper feedback circuit 

As previously indicated, the load applied at the stylus has but little 
effect upon the performance of the system as long as the enumerated 

FIG. 4. Feedback recorder in place on a commercial disk-recording machine. 

requirements are met. It is therefore interesting to note that the re- 
sponse measured with the stylus cutting a commercial direct record- 
ing material is, for practical purposes, identical with that measured 
with the stylus vibrating in free air. 

Since the feedback coil is rigidly coupled to the stylus, the voltage 
induced by its motion is proportional to the stylus velocity. The 
feedback voltage, properly amplified, therefore, provides means for 
easy field calibration of the device and for monitoring purposes. 

By virtue of the feedback feature, distortion and noise products 
created within the system are suppressed by the same amount as 

Jan., 1938] 



fundamentals of the same frequency. A harmonic analysis of the ve- 
locity of the stylus at 300 cps. while cutting commercial nitrocel- 
lulose direct-recording material showed second and third harmonics 
36 and 43 db., respectively, below the fundamental. 

The feedback feature of the recorder eliminates the critical ele- 
ments found in earlier types of recorders, and the problem of careful 
adjustment and maintenance has been eliminated. The slight varia- 
tions anticipated in the manufacture of the commercial product have 
no appreciable effect upon the ultimate characteristics, and, as a re- 
sult, high uniformity of performance is anticipated. 

L< __rv)NC'<<>'** 



f ' 













* \ 








100 200 500 1000 2000 5000 


10,000 20,000 

FIG. 5. Curves showing the stylus velocity for a constant signal 
input to the recorder amplifier system. (^4) Without feedback. 
(B) With properly controlled feedback. 

The device pictured in Fig. 4 has been subjected to an extensive 
field trial and has met all expectations both as to mechanical relia- 
bility and technical performance, and recordings made with it have a 
tone clarity not always attained with earlier recorders. Judged by 
the severest standards, the new recorder marks a definite forward 
step in the art of sound recording and warrants the consideration of 
the sound recording and reproducing industry. 


1 FREDERICK, H. A.: "Vertical Sound Records: Recent Fundamental Ad- 
vances in Mechanical Recordings on 'Wax,'" /. Soc. Mot. Pict. Eng., XVIII 
(Feb., 1932), No. 2, p. 141. 

1 NYQUIST, H.: "Regeneration Theory," Bell Syst. Tech. J., XI (Jan., 1932), 
No. 1, p. 126. 


3 BLACK, H. C.: "Stabilized Feedback Amplifiers," Bell Syst. Tech. J., XIII 
(Jan.. 1934), No. 1, p. 1. 


MR. KELLOGG: How much of the voltage developed in the exploration coil is 
due to motion and how much to direct induction? Also, what does the voltage 
due to direct induction, do to the characteristic? 

MR. WIEBUSCH : If we depended solely upon the induction, we should get a 
rather poor characteristic. The voltage due to direct induction must be small 
compared to the voltage generated due to the velocity, even at frequencies some- 
what beyond the range of interest. This is one of the major design problems. At 
10,000 cps., the voltage due to direct induction may be 5 or 10 per cent of that 
due to velocity. In the middle range it is, of course, very small perhaps a small 
fraction of one per cent. 


Summary. A brief outline of a photographic method of "blooping," devised to 
overcome the disadvantages of the previously used patching methods. 

In the first several months during which sound-track work was 
done in this laboratory, we became acutely aware that the prevailing 
methods of "blooping" or silencing the noise made by splicing two 
pieces of film presented a definite problem. All the conventional 
methods in use, while partially effective, were not quite satisfactory 
due to steadily increasing improvements in sound recording and re- 
producing which tended to magnify the disturbances caused by the 

The most common blooping patch in use was the triangular cut-out, 
which consisted merely in punching out a triangular portion of the 
negative at the position of the splice. The cut-out is one-half inch 
wide at the base and approximately one-eighth inch long. Although 
the "bloop" is reduced by this method, an objectionable sound still 
remains due particularly to the recent increased use of push-pull 

The first deviation from the conventional triangular cut-out is 
what is known as the half -moon bloop, increasing the length to five- 
eighths of an inch, and thereby improving the reduction of noise. 
Further improvement is attained by making the triangular shape 
still longer, this cut-out being three-quarters of an inch in length 
and intercepting the bias line at a smaller angle. 

In our experiments on blooping we discovered that the means of 
making the cut was as important in reducing the noise as the angle or 
the shape of the cut-out. A dull knife caused ragged edges on the 
film, and failure to keep the film flat at the time of cutting adds com- 
plications in the way of scratches, scars, etc. 

* Presented at the Spring, 1937, Meeting at Hollywood, Calif. ; received June 
11, 1937. 

** Williams Laboratories, Hollywood, Calif. 



A new blooper was built that much more accurately aligned the 
knife and also included a foot for holding the film firmly during the 
cutting. The new blade was set at an angle so as to shear the bloop. 
Although these improvements resulted in a decided reduction of 
noise, we were still not satisfied with the results. 

A more satisfactory method of blooping consists in spraying ink 
from an air-brush, through a mask, upon the film. The method re- 
quires much time and skill in application, and also invariably tends 
to peel off the film and cause noisy tracks. However, graduating the 
densities to the bloop, as is done in this method, did eradicate the 
noise of the splice, so the problem was to create another and satisfactory 
method of achieving the same results. 

The solution of the problem was a photographic method of graduat- 
ing the densities. A light is placed on the printer opposite the back 
side of the raw stock. The film is notched five inches ahead of the 
splice, causing the light to flash as the splice passes. The change of 
thickness of the film at the splice may also be used for flashing the 
light. The flash causes a fogged spot, similar to the spot produced by 
the air-brush, resulting in a quiet bloop, and eradicating all possibility 
of dirt or variation in the size and shape of the area. Furthermore, 
the speed of the operation has been greatly increased, and the method 
is simple and consistently accurate. 


During the Conventions of the Society, symposiums on new motion picture appara- 
tus and materials are held in which various manufacturers of equipment describe and 
demonstrate their new products and developments. Some of this equipment is de- 
scribed in the following pages; the remainder will be published in subsequent issues 
of the Journal. 


O. B. DEPUE** 

The non-slip principle has been comprehensively discussed by E. W. Kellogg 
in a previous issue of the JOURNAL. l In printers employing this principle the ex- 
posure is made while the films are in contact on an idle roller located several inches 

FIG. 1. Friction coupling, through whose action the 
motors are allowed to come into step. 

from the film-driving sprocket. If, instead, printing is done on a sprocket, con- 
tinual slipping occurs, with consequent detriment to the printed image. 

In the earlier Depue continuous printer the film was moved through the two 
printing stations by sprockets having positive gear drives. Power was supplied 

* Presented at the Spring, 1937, Meeting at Hollywood, Calif. ; received May 
17, 1937. 

** O. B. Depue, Inc., Chicago, 111. 




by a 1 /4-hp. motor to a speed-reducing worm and gear unit which, in turn, drove 
the vertical main shaft. The feed, take-up, and printing sprockets were driven 
from this line shaft by right and left helical gears. The drive was positive and 
more uniform than one involving the use of a belt, and did not require a flywheel. 
About two years ago it was discovered through stroboscopic observation that 
this drive, previously regarded as satisfactory, was far from perfect. Gear im- 
perfections were found to cause most of the variation in motion, although the 
clutch and couplings contributed their share. 

FIG. 2. Front view of printer showing film path. 

In the search for constant speed, it was found that a synchronous motor with 
built-in worm and gear reduction furnished a very steady and accurate drive of 
the desired speed, and permitted direct coupling to the printing sprocket shaft. 
This coupling consisted of a rather heavy soft-rubber disk, 5 inches in diameter 
and 3 / 8 inch in thickness, between two metal disks one of which was on the motor 
shaft and one on the sprocket shaft. Thus positive drive was attained with a 
vibration-absorbing coupling. This direct drive was later applied to the sound- 
printing unit, and the picture-printing sprocket was driven through a shaft and 
gearing. The picture sprocket showed the familiar unevenness of motion when 

Jan., 1938] 



viewed with the stroboscope. This arrangement was rejected because it failed to 
improve both printing drives. In the face of a demand for higher quality work in 
general, it would not have been sufficient to improve the sound printing only. 

Finally, it was concluded that the film must have an independent constant-speed 
drive for each printing station. In order to maintain synchronism between the 
two printing sprockets during starting and stopping, it was necessary to provide a 
linkage between the respective driving units. At first this was attempted by 

FIG. 3. Rear view of printer. 

means of a positive drive connection through gearing. This resulted in intolerable 
instability of motor performance. With the mechanical connections employed, 
an angular displacement of the apposite field centers of one motor or the other 
occurred, with the results that only one motor could run properly at any time. 
The difficulty was happily cured by fitting to one of the reduction-gear shafts a 
friction coupling comprising an adjustable spring bronze spider with an adjusting 
nut and check-nut. This allowed the motors to slip instantly into step, and they 
remained so throughout the cycle of operation (Fig. 1). The stroboscope showed 
the device to be thoroughly dependable. It was necessary to limit the tension so 
as not to defeat the purpose of the coupling, but once adjusted by the nut it re- 
quired no more attention. The resulting smooth motion of the printing]sprockets 



[J. S. M. p. E. 

not only provided higher-quality sound, but an improvement in the picture print 
as well. Kellogg, in the paper referred to previously, discussed and illustrated 
the non-slip and ultraviolet illumination features as well as the stabilizer utilized 
in connection with the picture-printing unit just described. 

Figs. 2 and 3 show how the two printing units are located, the picture unit above 
and the sound below. It is common practice in printing to run the film in the 
forward direction relative to the subject matter, and to print the picture and 
sound on the same machine in one run. The resulting position of the sound-track 
in the printer called for special construction of the optical path of the printing 
illuminator to allow locating the oil-encased flywheel unit or stabilizer on the back 
of the main plate. The illuminator optics had to be mounted on the front of the 


FIG. 4. Diagram of ultraviolet non-slip sound printer. 

main plate in such a way as not to interfere with threading (Fig. 2) . The illumi- 
nating slit image had to reach the negative emulsion at the point of contact with 
the positive film (Fig. 4) and had to be projected outward from a point near the 
axis of the constant-speed drum that supported the negative during printing. At 
this point the positive film was held in contact with the negative by means of a 
flanged pressure-roller. To meet these requirements the illuminating slit image 
was formed by two */ 4-inch right-angled prisms on a mounting extending under 
the overhanging portion of the constant-speed drum. The remainder of the il- 
luminating system lay upon an axis displaced such a distance from the edge of 
the film as to permit easy threading of the sound printing station drum. 

The light-source is a 7V2-ampere, 10-volt exciter lamp, supplied by a 10-ampere, 
10-volt d-c. motor-generator. Lamp voltage is adjusted by means of a 35-ohm 
circular rheostat in the generator field circuit. The control knob is located on the 
upper part of the printer, just above the meter. 

Jan., 1938] 



In the optical system is an adjustable mechanical slit 0.010 inch wide. The 
Bausch & Lomb objective has a focal length of 32 mm. Light of wavelength 
longer than the required ultraviolet is filtered out by means of a Corning No. 584 
ultraviolet transmitting glass filter, 5 /s inch in diameter and 0.045 inch in thick- 
ness. By employing 6.8 amperes with this arrangement of lamp, slit, and prisms, 
sufficient exposure is obtained for variable- width records. It is possible to use an 
automatic light-control device in the lamp circuit by utilizing the scene-end switch 
located in the idler unit of the sound feed-out sprocket at the right and immedi- 
ately above the printing unit. Lamp life is extended by using less than the full 


f 2.30 


if J.JO 


FIG. 5. Overall frequency response of printer. 

rated voltage. A 200-watt, 100-volt projection lamp has been employed by at 
least one user but has developed excessive heat, requiring two forced-ventilation 
units for cooling. Such a lamp having two vertical filament coils is inefficient for 
illuminating a horizontal slit. 

The overall response characteristic (Fig. 5) obtained with this non-slip printing 
device and ultraviolet illumination compared to that of the sprocket printer shows 
that both are uniform up to 3000 cps. Then the characteristic obtained with the 
sprocket printer falls off approximately 1 db. for each increase of 1000 cps., reach- 
ing a total loss of 6 db. at 10,000 cps. The non-slip printer yields a generally 
straight line except for an increase of 1 db. between 5000 and 9000 cps. (due, 
possibly, to the recording), followed by a decrease at 10,000 cps. of but slightly 
more than 1 db. The sprocket-wheel print shown in the chart was made on this 
same printer, but through an ultraviolet glass 0.045 inch thick. 


1 KELLOGG, E. W. : "A Review of the Quest for Constant Speed," /. Soc. Mot. 
Pict. Eng., XXVIH (April, 1937), No. 4, p. 337. 



[J. S. M. P. E. 


The primary consideration in the design of the Cine Kodak Model R (Fig. 1) 
was simplicity of operation and control. To obtain this simplicity, a single-plane 
film path with the supply reel above and ahead of the take-up was adopted. This 
retains the easy threading of a vertical camera with the added advantage 

FIG. 1. Cine Kodak model E. 

of greater stability (Fig. 2). The resulting form resembles the Cine Kodak Spe- 
cial with the 200-ft. film chamber. One of the principle advantages of this shape 
is the fact that the camera can be used without interference with the brim of a hat. 
The mechanism is built as a unit, with all controls mounted on the mechanism 

* Received Feb. 18, 1937. 
** Eastman Kodak Co., Rochester, N. Y. 


frame (Fig. 3). The spring motor, which is wound by a large key, pulls about 
18 feet of film per wind at I 1 /* feet per revolution. The rotating disk shutter, 
with a 165-degree opening, is driven by a pair of spiral bevel gears. The pull- 
down claw is a single formed steel piece driven by an eccentric and guided by a 
fixed stud with a cam surface on the claw element. The camera operates at three 
speeds: 16, 32, and 64 pictures per second, controlled by a variable-speed gov- 
ernor running at 2.4 times the speed of the pull-down. The governor weights 
act upon the disk through a cam surface, so the relation between the weights and 
the spring changes automatically and correctly with speed. The camera can be 

FIG. 2. View showing magazines and film path. 

operated with the trigger half-way down, from which position it will return when 
released,'or the trigger can be locked in the running position to allow the operator 
to get into the picture. 

The lens support, shutter housing, and film-track are combined into a unit, 
rigidly mounted upon the mechanism frame. This makes it possible to disas- 
semble the camera completely without disturbing the focus or the alignment of 
the lens. Both aperture plate and pressure pad are relieved to avoid damage to 
the picture area. The pressure pad withdraws the claw from the film path when 
it is moved back to admit the film. The pressure pad is easily removable to per- 
mit cleaning the gate. 

Despite the fact that the camera is in the "inexpensive" price group, the same 
standards of accuracy at important points are maintained in production as with 
all other Cine Kodaks. All gear centers are bored and reamed in a single sub- 
stantial fixture. Gears are generated and checked, shafts are turned to a toler- 


ance of 0.0005 inch and burnished, and mechanisms are "run in" before final timing 
and checking. 

The case contains the finder and film meters. The finder system is built 
through the case in such a position as to have extremely short parallax. A fea- 
ture of the camera is the addition of a supplementary film meter scale adjacent 
to the field of the finder. 

FIG. 3. The mechanism. 

Standard equipment is an f/3.5 20-mm. fixed-focus Kodak anastigmat lens in 
a standard threaded mount. Additional external features are the conventional 
self-setting film meter, a simplified etched exposure guide, tripod nut, and a 
carrying handle. A safety guard on the cover prevents closing the camera with 
the sprocket guards open. The camera can be used on the Cine Kodak titler or 
on a tripod. Without lens the camera is 8 inches long, 6 3 /4 inches high, and 2*/4 
thick. It weighs approximately 5 pounds when loaded. It will take standard 
50- or 100-ft. 16-mm. reels. 


The use of blimps to house motion picture cameras for making sound pictures 
places a burden upon the cameraman in that he can not be heard outside the 
blimp when directing the line-up of a scene. This was realized some years ago, 
and attempts were made at that time to overcome this obstacle. W. Daniels, 
cameraman at Metro-Goldwyn-Mayer Studios at that time, mounted an amplifier 
upon the platform of his rotambulator and placed a microphone inside the blimp. 

* Received Oct. 9, 1937. 

**Metro-Goldwyn-Mayer Studios, Culver City, Calif. 

t Los Angeles City Schools, Los Angeles. Calif 

Jan., 1938] 



This enabled the cameraman to direct the placement of lights and establish the 
movement in the scene without removing his eye from the eyepiece or his head 
from the blimp. 

The amplifier answered the purpose, but obviously was quite bulky, and the 
need of a well designed amplifier was evident. Analysis of the situation showed 
the following requirements of such an amplifier: 

(1) It must operate on 110 volts d-c. or a-c. continuously for twelve hours 
at a time. 

(2} It must have adequate power and distribution to cover a whole sound 

(3) Its frequency range should cover the voice range. 


FIG. 1. Circuit diagram of the amplifier. 

(4) Distortion should be at a minimum. 

(5) It should have enough gain to operate to the point of feedback when the 
gain-control is 90 to 100 per cent open and the microphone in the blimp. 

(6) It should be automatically turned off when the blimp door is closed. 

(7) A switch should be provided for changing the polarity of the supply 
voltage at the blimp. 

(8} Visual means should be available of knowing when the amplifier is ready 
for operation. 

(9) A protective fuse should be provided. 

(10) It should be insulated from ground throughout. 

(11) Complete operation without batteries should be possible. 

(12) Adequate filtering should reduce the a-c. hum to a minimum. 



(13) Extra switch should be provided to enable the amplifier to be operated 
without the automatic switch. 

(14) It should be compact, sturdy, and rigid. 

(75) Single-bracket mounting should be arranged, without interference with 
light-mounting brackets. 

(16) Storage space should be available for all cable tied to the amplifier. 

(17) Automatic switch, reversing polarity switch, and tone switch should be 
mounted hi a convenient position near the blimp door and the operator. 

(18) The microphone should be insulated. 

(19) Microphone mounting should be adequate. 

(20) The design should be standard. 

(21) It should harmonize with surrounding equipment. 

(22) It should be accessible for servicing. 

FIG. 2. The complete unit. 

The introduction of the 25L6 tube made possible a design of amplifier that 
could meet in all respects the demands listed above. Fig. 1 shows the circuit 
used in the amplifier, making use of the cathode phase-inversion circuit using a 
6J7 to drive two 25L6's in push-pull. A 25Z6 single-phase rectifier operating 
into a condenser input filter is used. The overall gain of the amplifier is sufficient 
to give full output from the 25L6's under normal operating conditions when the 
gain is 75 per cent open. The point of feedback is usually reached when the 
gain is totally on and the microphone is in position in the blimp. No more gain 
is needed, and it was found that the annoyance factor is practically minimized 
by providing only sufficient gain to eliminate feedback at all times. The circuit 
harmonic distortion is less than 5 per cent. The acoustical characteristics have 
some distortion due to cabinet resonance, but the distortion is not objectionable; 


in fact, it gives to the output a character that enables the amplifier to be heard 
over normal production noises. 

A six-inch dynamic speaker provides adequate coverage for the average set. 
Screened openings in the back of the cabinet allow for some sound distribution 
behind the camera as well as serving for ventilation of the unit. 

The unit is operated completely on a 110-volt d-c. or a-c. circuit. The current 
is fed through a switch that turns on the amplifier when the blimp door is open. 
A polarity reversing switch enables the cameraman to change the supply polarity 
at the blimp when operated by direct current. A switch operating a buzzer or 
feedback circuit producing an amplified tone in the speaker is also built into the 
switch-box. The tone is used to attract attention upon the set when the camera- 
man has completed his work. This type of signal has been found to reduce 
confusion upon the set and to save production time. 

FIG. 3. Cabinet and chassis, opened. 

A pilot-light is operated from the rectified current, and is used to notify the 
operator when the unit is operating and when the d-c. supply is properly polarized. 
Due to the fact that a separate voltage is supplied to the blimp to operate the 
camera, and in some cases one side of this camera supply voltage is grounded, 
it is necessary to insulate the amplifier and all other parts carefully. 

Sometimes the amplifier unit is used with silent shots. In such cases the blimp 
may not be used and an additional switch is then used with the amplifier. 

The two-button carbon microphone is suspended from the camera eyepiece, 
conveniently located for the cameraman when looking through the lens. 

The cabinet and chassis are made of a 20-gauge body steel with reinforced 
corners, and is 6 l /2 inches square and 7 inches high. A compartment to hold the 
cables and microphone is built in the upper part of the cabinet. All resistors 
and condensers are mounted rigidly upon a strip of formica. A volume control, 


fuse holder, pilot light, and a four-prong adapter are mounted upon the rear 
panel below the cable compartment. A metal grill protects the loud speaker 
on the front end of the cabinet. A single socket to receive a x /< inch by 1 inch 
iron bracket is used to hold the amplifier alongside the blimp. The cabinet and 
switch -box are finished in gray crackle paint with chrome hardware. Access to 
the tubes and inside the cabinet is accomplished by removing the rear panel and 
chassis. Sheet metal screws hold the panel and chassis in place. 

The number of different kinds of blimps in use makes it difficult to standardize 
upon the switching and mounting methods. It has been found that in most 
major studios there is enough standardization of camera equipment to design 
a switch-box and mounting bracket for these studios and use standardized 
amplifier units in all cases. 

Cameramen have estimated that from forty-five minutes to an hour and a 
half a day are saved in production tune by these camera amplifiers. Directors 
have remarked that less confusion and noise is experienced upon the set due to 
the fact that the cameraman's instructions can be heard without repetition or 
excessive shouting. 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in those magazines that are available may be obtained from the Library of the U. S. 
Department of Agriculture, Washington, D. C. 

American Cinematographer 

18 (Sept., 1937), No. 9 
Reaction on Making His First Color Production (p. 

408). J. W. HOWE 

Cinematography with 8-Mm. Cameras Aid in Den- 
tistry (p. 426). H. A. LINEK 
Three-Lens Turret Built for 8-Mm. Users (p. 429). 
Automatic Development and Its Advantages (p. 435). 

Bell Laboratories Record 

16 (Oct., 1937), No. 2 

Diphonic Loud Speaker for Mirrophonic Sound Sys- 
tems (p. 53). R. C. MINER 

Limitations in High-Frequency Band-Filter Design 

(p. 56). C. E. LANE 


17 (Oct., 1937), No. 10 

Television Economics (p. 10). A. N. GOLDSMITH 

Linear Amplifier Adjustments (p. 11). A. J. EBEL 

A Simplified Theory of Filter Selectivity (p. 12). H. DUDLEY 

Disk Recording Equipment and Its Quality Require- 
ments (p. 17). T. L. DOWEY 
Plate Resistance Control in Vacuum Tubes as Audio 

Gain Control Means (p. 23). A. W. BARBER 

Educational Screen 

16 (Oct., 1937), No. 8 
The Motion Picture as an Aid to Learning (p. 252) W. M. GREGORY 


10 (Oct., 1937), No. 10 

Scanning in Television Receivers (p. 18). F. J. SOMERS 

Quality in Disk Reproduction (p. 25). C. J. LEBEL 

Reactance Amplifiers (p. 28). 

Television in Great Britain (p. 32). H. M. LEWIS AND 



120 CURRENT LITERATURE [j. s. M. p. E. 


13 (Oct., 1937), No. 14 
Regelschnellschreiber, ein neues Messgerat (Regel Rapid 

Recorder, a New Measuring Device) (p. 161). P. HATSCHEK 

Fernseh-Filmabtastgerat (Television Scanner) (p. 163). SCHRIEWER 


10 (Oct., 1937), No. 10 
Commercial Cathode-Ray Tubes (p. 267) R. R. BATCHER 

International Photographer 

9 (Oct., 1937), No. 9 

Light Meters and Color (p. 11). T. S. CURTIS 

Superpan and Infrared (p. 14). H. MEYER 

Ready Playback Recordings (p. 36). J. N. A. HAWKINS, 

Albin Indicator (p. 37). 

International Projectionist 

12 (Oct., 1937), No. 10 
An Analysis of Imperfections Apparent on the Screen 

(p. 7). A. C. SCHROEDER 

More Data on the W. E. Mirrophonic Speaker System 

(p. 20). R. C. MINER 

Typical Troubles in Modern Sound Reproducing Units 

(p. 24). L. CHADBOURNE 

Journal of the Acoustical Society of America 

9 (Oct., 1937), No. 2 
Factors in the Production of Aural Harmonics and 

Combination Tones (p. 107). E. B. NEWMAN, 


A New Interpretation of the Results of Experiments 
on the Differential Pitch Sensitivity of the Ear (p. 
129). W. E. KOCK 

A Method for Evaluating Compliant Materials in 

Terms of Their Ability to Isolate Vibrations (p. 141) . W. JACK AND 


Various Types of Absolute Pitch (p. 146). A. BACHEM 

A Direct-Reading Pitch Recorder and Its Applications 

to Music and Speech (p. 156). J. OBATA AND 

Sound-Waves in a Moving Medium (p. 162). J. D. TRIMMER 

Journal of the Optical Society of America 

27 (Oct., 1937), No. 10 

Orthostereoscopy (p. 323). H. F. KURTZ 

An Examination of the Principles of Orthostereoscopic 

Photomicrography and Some Applications (p. 340). L. C. MARTIN AND 



Journal of Scientific Instruments 

14 (Oct., 1937), No. 10 

High-Gain Low-Frequency Amplifiers (p. 325). A. F. RAWDON-SMITH 

A Note on the Calibration of Audio-Frequency Oscil- 
lators (p. 339). N. F. ASTBURY 


19 (Sept., 1937), No. 10 

Die Eurocord-Optik (Eurocord Optics) (p. 226). K. SCHWARZ 

Die Photographic im Dienste der Dokumentation 

(Photography in Documentation Service) (p. 229). O. FRANK 
Uber widerstandslosen Bogenlampenbetrieb (Low Re- 
sistance Arc Lamp Operation) (p. 230). H. TUMMEL 
Der Lautstarkenumfang der Eurocord-Schrift (Sound- 
Intensity Range of the Eurocord Recorder) (p. 232). H. WOHLRAB 

Motion Picture Herald (Better Theaters Section) 

129 (Oct. 16, 1937), No. 3 

Theatre Acoustics Today (6. Mounting and Decorat- 
ing Acoustic Materials) (p. 61). C. C. Pox WIN 

Proceedings of the Institute of Radio Engineers 

25 (Nov., 1937), No.ll 
A Low-Distortion Audio-Frequency Oscillator (p. 

1387). H. J. REICH 

Review of Scientific Instruments 

8 (Oct., 1937), No. 10 

Two Simplified Technics for Synchronized X-Ray, 
Sound Recording, and Cathode-Ray Oscillographic 
Studies of Speech (p. 382). R. CURRY 




Officers and Committees in Charge 

W. C. KUNZMANN, Convention Vice-President 
J. I. CRABTREE, Editorial V ice-President 
G. E. MATTHEWS, Chairman, Papers Committee 
W. WHITMORE, Chairman, Publicity Committee 
E. R. GEIB, Chairman, Membership Committee 

Local Arrangements and Reception Committee 






N. D. GOLDEN, Chairman 

Registration and Information 

W. C. KUNZMANN, Chairman 

Ladies' Reception Committee 

MRS. R. EVANS, Hostess 

assisted by 

Banquet Committee 

R. EVANS, Chairman 

Publicity Committee 

W. WHITMORE, Chairman 










P. A. McGuiRE 




Convention Projection Committee 

H. GRIFFIN, Chairman 



Officers and Members of Washington Projectionist Local 224. 

Membership Committee 

E. R. GEIB, Chairman 


Hotel and Transportation Committee 

J. G. BRADLEY, Chairman 




The headquarters of the Convention will be the Wardman Park Hotel, where 
excellent accommodations are assured. A reception suite will be provided for the 
ladies, for whom also is to be arranged an interesting program of entertainment. 

The following daily hotel rates, European plan, are guaranteed to SMPE 
delegates attending the Convention : 

One person, room and bath $ 3.50 

Two persons, standard bed 5 . 00 

Two persons, twin beds 5 . 00 

Parlor suite, one person 9 . 00 

Parlor suite, two persons 11 . 00 

Room reservation cards will be mailed to the membership of the Society in the 
near future, and those who plan to attend the Spring Convention should return 
their cards promptly to the Wardman Park Hotel to be assured satisfactory 
accommodations. Local railroad ticket agents should be consulted with regard to 
trains and rates. 

Technical Sessions 

An attractive and interesting program of technical papers is being assembled 
by the Papers Committee. All technical sessions, apparatus symposiums, and 
film programs will be held in the Little Theatre of the Hotel. 

Registration and Information 

The Convention registration headquarters will be located at the entrance of 
the Little Theatre, where all the technical sessions will be held. The members of 
the Society and guests attending the Convention are expected to register and 
receive their badges and identification cards for admittance to special evening 
sessions. These cards will also be honored at several de luxe motion picture 
theaters in Washington during the four days of the Convention. 


Informal Luncheon and Semi- Annual Banquet 

The usual informal Luncheon will be held at noon of the opening day of the 
Convention, April 25th, in the Continental Room of the Hotel. On the evening 
of Wednesday, April 27th, will be held the Semi-Annual Banquet of the Society, 
in the Continental Room of the Hotel at 7:30 P.M. Addresses will be delivered 
by prominent members of the industry, followed by dancing and other entertain- 

Points of Interest 

To list all the points of interest in and about Washington would require too 
much space, but among them may be mentioned the various governmental 
buildings, such as the Capitol, the White House, Library of Congress, Department 
of Commerce, U. S. Treasury, U. S. Bureau of Standards, Department of Justice, 
Archives Building; and other institutions such as the National Academy of 
Sciences, the Smithsonian Institution, George Washington University, Washing- 
ton Cathedral, Georgetown University, etc. In addition may be included the 
Lincoln Memorial, the Washington Monument, Rock Creek Park, The Francis 
Scott Key Memorial Bridge, Arlington Memorial Bridge, the Potomac River, 
and Tidal Basin. Mt. Vernon, birthplace of Washington, is but a short distance 
away and many other side trips may be made conveniently via the many highways 
radiating from Washington. 


The Wardman Park Hotel management is arranging for golfing privileges for 
SMPE delegates at several courses in the neighborhood. Regulation tennis 
courts are located upon the Hotel property, and riding stables are within a short 
distance of the Hotel. Trips may be arranged to the many points of interest in 
and about Washington. 



As a result of the recent elections, the Officers and Managers of the three Local 
Sections of the Society for the year 1938 are as follows: 

Atlantic Coast Section 

*G. FRIEDL, JR., Chairman 

L. W. DAVEE, Past- Chairman *H. GRIFFIN, Manager 

*D. E. HYNDMAN, Sec.-Treas. **P. J. LARSEN, Manager 

Mid-West Section 

*S. A. LUKES, Chairman 

C. H. STONE, Past- Chairman *B. E. STECHBART, Manager 

*A. SHAPIRO, Sec.-Treas. **G. W. BAKER, Manager 

Pacific Coast Section 

*J. O. AALBERG, Chairman 

K. F. MORGAN, Past-Chairman *H. W. MOYSE, Manager 

*G. A. CHAMBERS, Sec.-Treas. **C. W. HANDLEY, Manager 


At a meeting held at the Studios of RCA Photophone, Inc., 411 Fifth Ave., 
New York, N. Y., on December 8th, a paper was presented by Mr. P. Arnold 
of the Agfa Ansco Corp., Binghamton, N. Y., entitled "Agfa Ultra-Speed Pan- 
chromatic Negative." Following the technical presentation several reels of film 
were projected, showing the photographic results attained with the new ultra- 
speed film as compared with shots made on the super-pan. 

About 250 persons attended the meeting and considerable interest in the 
subject was shown by the extended discussion following the presentation. 


Messrs. M. Townsley, L. B. Hoffman, and P. Foote, all of the Bell & Ho well 
Engineering Laboratory, were the participants in a symposium on the subject 
of "High-Speed Photography," held on December 16th in the auditorium of the 
Bell & Howell Engineering Laboratory, Chicago. 

The presentations were accompanied by motion pictures demonstrating the 
application of high-speed photography to the design and manufacture of appa- 

* Term expires December 31, 1938. 
** Term expires December 31, 1939. 



ratus in engineering and commercial fields. The Bell Telephone Laboratory 
film, shown at the recent New York Convention by W. Herriott, was also in- 

The next meeting of the Section will be held on Tuesday, January 11, 1938. 


Four outstanding papers delivered at the recent Fall Convention of the Society 
in New York, selected as of particular interest to members in the Hollywood area, 
formed the program of the meeting of the Pacific Coast Section, held on De- 
cember 9th at the Sunset Arbor Cafe Auditorium in Hollywood. The four 
papers were: 

"Film Perforation and 96-Cycle Frequency Modulation in Sound-on-Film," 
by J. Crabtree and W. Herriott, Bell Telephone Laboratories, New York, N. Y. 

"High-Speed Motion Picture Photography Applied to Design of Telephone 
Apparatus," by W. Herriott, Bell Telephone Laboratories, New York, N. Y. 

"Modulated High-Frequency Recording as a Means of Determining Conditions 
for Optimal Processing," by J. O. Baker and D. H. Robinson, RCA Manu- 
facturing Co., Camden, N. J. 

"Reduction of Loop-Length Variations in Non-Slip Printers," E. W. Kellogg, 
RCA Manufacturing Co., Camden, N. J. 

The meeting was well attended and considerable interest was shown in the 
presentation. The papers were read by several members of the Section. 


At a recent meeting of the Admissions Committee, at the General Office of 
the Society, the following applicants for membership were admitted to the 
Associate grade: 

DORN, H. P. KAY, H. 

9007 Detroit Ave., 1075 Grand Concourse, 

Cleveland, Ohio. Bronx, N. Y. 


973 Dumont Ave., 5141 Ellis Ave., 

Brooklyn, N. Y. Chicago, 111. 


556 Chelton Ave., 6 

Camden, N. J. New 


Indian Motion Picture KOSSOWSKY, A. 

Producers' Assn., RCA Victor Co 

Bombay, India. Buenos Alres ' 


1028 New York Ave., Progressive Education Assn 

Brooklyn, N. Y. 160 Broadway, 

HOPKINSON, K. New York - N - Y 

9 Kenilworth St., MOTELOW, H. J. 

New South Wales, 653 Hendrix St., 

Australia. Brooklyn, N. Y. 



Core Cinesound Productions, Ltd., Bane of Boston, 

New South Wales, Buenos Aires, Argentina. 

Australia. TERHUNE, H. J. 

SANSONE, M. Sea Island Beach, 

3440 Fish Ave., P. O. Box 876, 

Bronx, N. Y. Georgia. 


1931 E. 47th St., Sydney, 

Brooklyn, N. Y. New South Wales, Australia. 


Warner Bros. Pictures, 7655 Sunset Blvd., 

Burbank, Calif. Hollywood, Calif. 

33 Cooper St., 
Brooklyn, N. Y. 

In addition, the following applicants have been transferred by vote of the 
Board of Governors to the Fellow and Active grades: 


25 Hunter Ave., RCA Manufacturing Co., Inc. 

Fanwood, N. J. Camden, N. J. 

S. M. P. E. 


These films have been prepared under the supervision of the Projection 
Practice Committee of the Society of Motion Picture Engineers, and are 
designed to be used as precision instruments in theaters, review rooms, 
exchanges, laboratories, factories, and the like for testing the perform- 
ance of projectors. 

Only complete reels, as described below, are available (no short sections 
or single frequencies). The prices given include shipping charges to all 
points within the United States ; shipping charges to other countries are 

35 -Mm. Sound- Film 

Approximately 500 feet long, consisting of recordings of several speak- 
ing voices, piano, and orchestra; buzz-track; fixed frequencies for focus- 
ing sound optical system; fixed frequencies at constant level, for de- 
termining reproducer characteristics, frequency range, flutter, sound- 
track adjustment, 60- or 96-cycle modulation, etc. 

The recorded frequency range of the voice and music extends to 10,000 
cps. ; the constant-amplitude frequencies are in 15 steps from 50 cps. to 
10,000 cps. 

Price $37.50 each, including instructions. 

35-Mm. Visual Film 

Approximately 500 feet long, consisting of special targets with the aid 
of which travel-ghost, marginal and radial lens aberrations, definition, 
picture jump, and film weave may be detected and corrected. 

Price $37.50 each, including instructions. 

16-Mm. Sound-Film 

Approximately 400 feet long; contents identical to those of the 35- mm. 
sound-film, with the exception that the recorded frequency range ex- 
tends to 6000 cps., and the constant-amplitude frequencies are in 11 
steps from 50 cps. to 6000 cps. 

Price $25.00 each, including instructions. 

16-Mm. Visual Film 

An optical reduction of the 35-mm. visual test-film, identical as to 
contents and approximately 400 feet long. 
Price $25.00 each, including instructions. 








Volume XXX FEBRUARY, 1938 Number 2 



Demonstration of Stereophonic Recording with Motion Pic- 
tures J. P. MAXFIELD 131 

Reduction of Loop-Length Variations in Non-Slip Printers. . . 

E. W. KELLOGG 136 

A Recorder for Making Buzz-Track E. W. KELLOGG 150 

Push-Pull Recording J. K. HILLIARD 156 

Theoretical Notes on the Push-Pull Method of Recording 

Sound O. O. CECCARINI 162 

Twenty Years of Development of High-Frequency Cameras. . 

H. E. A. JOACHIM 169 
Grain Size Determination and Other Applications of the Callier 


Report of the Honorary Membership Committee to the Board 

of Governors ' 191 

Safeguarding and Developing Our Film Markets Abroad 


Hunting the Songs of Vanishing Birds with a Microphone 

Notes on the Procedure for Handling High-Volume Release 

Prints J. K. HILLIARD 209 

Academy Standard Fader Setting Instruction Leader 215 

New Motion Picture Apparatus 

A Mobile Sound Recording Channel 


A Simplified Device for Cueing Motion Picture Films 


Current Literature 229 

vSpring, 1938, Convention; Washington, D. C., April 25th-28th, 

Inclusive 232 

Society Announcements 236 

Constitution and By-Laws 238 





Board of Editors 
J. I. CRABTREE, Chairman 



Subscription to non-members, $8.00 per annum ; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West-Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1938, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 


^President: S. K. WOLF, 100 E. 42nd St., New York, N. Y. 

* Past-President: H. G. TASKER, Universal City, Calif. 

^Executive Vice-President: K. F. MORGAN, 6601 Romaine St., Los Angeles, 


^Engineering Vice-President: L. A. JONES, Kodak Park, Rochester, N. Y. 
^Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 
** Financial Vice-President: E. A. WILLIFORD, 30 E. 42nd St., New York, N. Y. 

* Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 
*Secretary: J. FRANK, JR., 90 Gold St., New York, N. Y. 

^Treasurer: L. W. DAVEE, 250 W. 57th St., New York, N. Y. 


*J. O. AALBERG, 157 S. Martel St., Los Angeles, Calif. 

*M. C. BATSEL, Front and Market Sts., Camden, N. J. 

**R. E. FARNHAM, Nela Park, Cleveland, Ohio. 

*G. FRIEDL, JR., 90 Gold St., New York N. Y. 

*A. N. GOLDSMITH, 444 Madison Ave., New York N. Y. 

**H. GRIFFIN, 90 Gold St., New York, N. Y. 

**A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

*S. A. LUKES, 6145 Glen wood Ave., Chicago, 111. 

*Term expires December 31, 1938. 

**Term expires December 31, 1939. 



Summary. As one of the events of the Fall, 1937, Convention at New York, N. Y., 
a demonstration of stereophonic sound with pictures was given in the auditorium of 
the Bell Telephone Laboratories. A special motion picture had been shot and re- 
corded with twin sound-tracks, which were reproduced through separate channels by 
loud speakers located at the sides of the screen. So far as is known, this was the first 
public demonstration of stereophonic sound in conjunction with motion pictures. 

Sound motion pictures, as presented today, are equipped with a 
single source of sound, a loud speaker usually placed centrally be- 
hind the screen. There is therefore no acoustic illusion of sound 
movement from one side of the screen to the other. As a result, our 
eyes have been trained to "pull" the sound the necessary distance 
sidewise, to make it appear to come from the visual image of its source. 
With stereophonic recording and reproducing, this mental strain is 
relieved, since the sound, of its own accord, moves back and forth 
across the screen to follow the image of its source. 

Stereophonic reproduction implies a localization of the apparent 
sound-source in both a sidewise and a fore-and-aft direction. This 
localization has been thoroughly accomplished. An unexpected ac- 
companiment to this localization is a marked improvement in the 
quality and in the sense of reality of the sound that is heard. From 
a commercial standpoint, the latter property is at least as important 
as the first. It is the one that is almost always commented upon by 
members of the general public, since their eyes have always helped 
them to obtain some apparent motion of the sound. 

In the demonstration presented here tonight, one should pay par- 
ticular attention to the location of .the apparent points, at or back of 
the screen, from which the various sounds appear to come, and should 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received Oc- 
tober 11, 1937. 

** Electrical Research Products, Inc., New York, N. Y. 


132 J. P. MAXFIELD [j. $. M. P. E. 

coordinate these points with the positions of the picture image from 
which the sounds are supposed to come. 

The illusion of position is very strong, and it is therefore necessary 
to correlate very carefully the apparent acoustic position with the 
visual image's actual position. 

In the early days of the development of this method, recording 
technics were not used, and the correlation between sound location 
and visual image was obtained by the use of pantomime artists on a 
real stage. A very amusing and almost ghostly effect was attained 
by having the pantomimer walk in a direction other than that in 
which the sound was moving. As the pantomimer got a little out 
of position, a very uncomfortable effect was produced, and suddenly 
the sound seemed to jump away from the pantomimer and follow 
its own course, leaving the pantomimer silent and useless. 

One can probably best appreciate the possibilities of this method 
by actually seeing and hearing a demonstration. The reel to be 
shown has been produced purely as an experiment, and is designed to 
show the engineering possibilities of stereophonies. We believe that 
these possibilities are so well demonstrated that those who are skilled 
in the arts will see the dramatic possibilities also. 

This development of stereophonic recording has formed a natural 
part of the general developmental work carried on for the purpose of 
improving the quality of talking pictures. Whether or not it will 
ever be adopted by the motion picture industry depends upon the 
motion picture producers. It is certain, however, that to obtain 
the full, ultimate illusion of reality it will be necessary to combine 
with a colored stereoscopic picture, stereophonic sound. 

At the close of these remarks by Mr. Maxfield, a specially pro- 
duced motion picture, with stereophonic recording, was projected 
and reproduced. The recording had been done with a four-ribbon 
light-valve, one pair of ribbons being actuated by current from the 
microphone at one side of the stage and the other pair by current 
from the microphone at the opposite side of the stage. The sound- 
tracks look somewhat like the tracks of the push-pull track, except 
that they are not recorded 180 degrees out of phase. The loud 
speakers were located behind the mask at either side of the center. 

The purpose of the picture was to afford opportunity for the spec- 
tators to localize the sounds being reproduced and to correlate the 
apparent sources of the sound with the images of the actors supposed 


to produce them. The first scene of the picture showed an orchestra 
of 40 or 50 players, the strings, brasses, tympani, etc., being located 
in the usual places. The nature of the musical selection was such 
as to permit noting from what part of the orchestra the sounds of the 
various instruments emanated. Among other effects the clashing 
of the cymbals was particularly noticeable as apparently arising from 
the upper left-hand corner of the screen, where the image of the per- 
former playing them was located. 

The next scene of the picture showed two men playing a game of 
table-tennis, or ping-pong. It was interesting to note how the sound 
of the impact of the ball upon the table travelled back and forth 
across the screen with the image of the ball. At one point, when the 
ball bounced from the table to the floor, the sounds of the successive 
impacts of the ball upon the floor could be followed as the ball bounced 
out of the picture. Conversation between the players at the same 
time demonstrated the manner in which the reproduction of the 
voices would seem to jump from one side of the screen to the other. 

The third scene of the picture started out with a black, or nearly 
black screen, the inside of a room at night, with the lights turned out. 
The sounds of an actor cautiously entering the room were heard; 
his collisions with furniture in the darkened room and his conversation 
with a companion could be heard as if the sounds were emanating 
from various portions of the screen as the actors moved about. 
Toward the end of the scene another actor entered the room and 
turned on the lights, permitting the audience to correlate the positions 
of the actors at the moment the light was turned on with their ac- 
tual positions upon the screen. The picture concluded with another 
scene of the orchestra. 


MR. PALMER: What does the sound-track of this film look like? 

MR. MAXFIELD: The sound-track looks like a push-pull track, except that 
the two tracks are not 180 degrees out of phase. It is made with a four-ribbon 
light- valve; one pair of ribbons is fed from the microphone on one side of the 
stage and the other pair from the microphone on the opposite side. 

MR. CRABTREE: I have been wondering why there are not more producing 
and exhibiting executives present this evening. It would seem to me that an 
exhibition of this nature is of vital importance to their future business. Un- 
fortunately these days it seems to be necessary for the engineer to hand out any 
new idea or invention in its finished form and on a golden platter before its po- 
tential applications can be fully appreciated. 

134 J. P. MAXFIELD [j. s. M. P. E. 

The industry owes a debt of gratitude to the Electrical Research Products, 
Inc., and, particularly to Vice-President Knox for this pioneering experiment, 
and the Papers Committee is especially proud to be able to give it a place of 
honor on our program. It was more than ten years ago that Dr. Steinberg 
demonstrated to me in the Bell Laboratories the astonishing dramatic effects 
attainable by binaural sound reproduction. Ever since, I have been looking 
forward to the time when such effects would be applied to the motion picture. 
Binaural reproduction, of course, requires the use of earphones and the dramatic 
effects which we have witnessed are by no means as impressive as the binaural 
effects but they are a step in the right direction. 

Novelty is what the film industry is lacking at the present time. I can go into 
a small theater today and see as good a show for 20 cents as I can for 75 cents or 
a dollar on Broadway. It would seem as if the large exhibitor should be looking 
for some novelty that would enable him to put on a better show than the little 

MR. MAXFIELD: With the stereophonic system we can not bring the sound 
out in front of the loud speakers, into the audience area. We can cover the 
whole area behind the screen from the face of the loud speakers back. 

MR. CRABTREE: What effect do you get by placing the speakers in the audi- 
torium and not behind the screen? 

MR. MAXFIELD: There is a minimum distance in front of the loud speakers at 
which we can work and not have the sound break up into two distinct sources. 
Moving the speakers out into the auditorium would force us to move the audience 
farther back and nothing would be gained. That is one of the limitations of 
the technic. 

One novelty would be to call attention to the position of the action on a dark- 
ened screen in some kind of play that calls for materializing a character or for the 
appearance of a ghost. The location of the sounds could be noted before the 
ghost is materialized. We can for the first time call visual attention to positions 
on the screen by the sound effect alone. The dramatic people can think of 
many things that could be done with such an effect. 

MR. RICHARDSON : If sounds are picked up at the sides of the stage and then 
brought out to the center, how would there be any difference? 

MR. MAXFIELD: If you are midway between the microphones, both micro- 
phones will get the same sound, and if these two similar sounds are reproduced 
from both sides of the screen they will appear as one sound coming from the 
center. Some of our theaters have loud speakers at the sides of the screen instead 
of at the middle, and the sound still seems to come from the middle of the screen 
because the same signal is fed into both loud speakers. If the actor moves a little 
to one side, the microphone on that side picks up a little more than the one at the 
other side. The complete explanation of that was published in the Bell Telephone 
Record two or three years ago by Fletcher and Steinberg. 

MR. FRANK: Have any studies been made to determine what effect the 
distance from the screen and the angle of vision will have upon the appreciation 
of stereophonic sound? 

MR. MAXFIELD: Only very rough determinations. At the present time it 
looks as if the distance of the audience from the screen will have to be of the 
order of one-half to two-thirds of the distance between the loud speakers. That 


means, for the front row of seats, about one-half to two-thirds of the width of 
the screen. The acoustics of the auditorium affect only the fore-and-aft il- 

MR. KELLOGG: How far back must one be in an auditorium such as this 
before he will fail to notice the difference between two-channel or single-channel 

MR. MAXFIELD: I have never been in a big enough auditorium to lose the 
effect. At the front you do not lose the two-channel effect, but you are con- 
scious that it is two-channel and not localized. At the back I have never lost 
the improvement in quality. The accuracy of localization from the back seats 
at the Philadelphia demonstration was a little better sidewise than localizing a 
performer on the darkened stage by listening to him. 

MR. KENDE: Is there a difference between the starting points of the two 

MR. MAXFIELD: No, the sound-tracks are side by side on the same film. 
They start at the same place and are reproduced with the double photocell. 

MR. CRABTREE: What is the relation between the fidelity of the reproducing 
equipment and the dramatic effect to be obtained, assuming the better the 
fidelity the better the dramatics? Would existing theaters require new equip- 
ment or could some of the equipment be adapted to this? 

MR. MAXFIELD: The theaters would have to equip themselves with two 
amplifier channels and two sets of loud speakers. Any statement of quality re- 
lationship is a matter of personal opinion. Personally, I would rather hear 
two-channel reproduction good to 6000 cps., than one-channel reproduction 
good to 15,000 cps. It is more pleasing, more realistic, more dramatic. I am 
not sure that all other engineers will agree with me. In discussions with other 
engineers I have obtained figures ranging from 8000 to 5000 cps., as being 
equivalent to a 15,000-cps. single channel. We shall not have to increase the 
frequency range in the theater to get a rather large dramatic improvement with 

MR. CRABTREE: What is the depth of field over which it is effective in a room 
of this size? 

MR. MAXFIELD: From about half or two-thirds the width of the screen, all 
the way back. 

MR. FRANK: In a very wide house, such as the Roxy Theater in New York, 
where the seats near the side are well off the center axis of the screen, what is the 

MR. MAXFIELD: The house in which we have done most of our work is the 
Academy of Music in Philadelphia, and there we covered the seats pretty well 
to the side. However, that is not an extremely wide house. We have not had 
experience in a theater like the Roxy. The experience we have had leads us to 
believe that as the picture begins to get a little out of shape, the reproduction 
will lose in accuracy of illusion; but even far around to the side, and at the 
front, we still have good quality. We merely lose accuracy of location. 



Summary. Compensation for varying degrees of film shrinkage is accomplished 
in the Bedford non-slip printer by changes in the length of a loop of film between a 
sprocket and the printing point. This involves uncertainty of synchronism by the 
amount that the loop, as first threaded, differs in length from the final running loop. 
For most purposes, the present designs do not cause greater change in loop-length 
than may readily be tolerated. 

For certain purposes, especially if this type of printer is to be employed for 16- 
mm. films, there may be too much departure from synchronism. A guide-roller ar- 
rangement is described by which the necessary change of angle of approach of the raw 
stock to the printing point is attained uith comparatively small change of loop length. 

Several possible arrangements are considered and other features of the non-slip 
printer are discussed. 

The principle of compensating for film shrinkage by automatically 
altering the curvature where a film is driven past an optical system, 
was first proposed by Bedford for application to projectors. 1 It 
might also be applied to recorders. The reason why it has not been 
applied in these fields is simply that other methods of accomplishing 
the purpose were available. As compared with use of a drum whose 
speed is controlled by the film, as for example, in magnetic- drive re- 
corders and rotary stabilizer sound heads, the Bedford principle has 
the drawback of calling for a drum of small diameter with the light- 
beam crossing the drum. This would not present an insurmountable 
difficulty but would give rise to awkward design problems. On the 
other hand, since Bedford's principle permits use of a fixed-speed 
drum, the drum may be driven through gears with suitable filtering, 
which is favorable when quick starting is important. 

An application in which this principle has found no competitors 
is sound printing. As both films are to be held against a drum, the 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received October 
11, 1937. 

** RCA Manufacturing Co., Camden, N. J. 



speed of the drum may be controlled by one of the films, but on 
practicable ways other than employing the principle of curvature 
compensation have been suggested for equalizing the surface veloci- 
ties of the negative and raw stock. Bedford early recognized that 
sound printers presented a problem to which his principle was pe- 
culiarly applicable, 2 and built a simple model of such a printer. It 
was not until several years later that active development of a printer 
of this type was taken up by C. N. Batsel 3 with cooperation and en- 
couragement of A. C. Blaney. In the meantime, Wood had indepen- 
dently conceived the idea and had built and published a description 4 
of a printer employing the identical principle. Such a delay in the 
utilization of many valuable ideas is typical of developments of this 
type. Engineers who are in a position to carry through such develop- 
ments to the commercial application stage may be under too great 
pressure to devote their attention to certain developments even 
though they may recognize their value. In this case, comparison 
of resolution attainable with a non-slip printer with that of a well 
adjusted sprocket printer did not fully reveal the advantages of the 
former, partly because of crudities in the experimental models and 
partly because at the time there were so many other causes of loss 
of resolution that the contribution of the printer to the losses was of 
secondary magnitude. Justification for developmental work under 
such circumstances may have to be based upon general principles, 
backed by the theorem that each element in a system must be made 
the best possible, even though other limitations may at the time be 
too serious to permit the benefit to appear. Subsequent refinements 
in recording, especially the advent of ultraviolet recording, have so 
far removed some of the other limitations in resolution that the 
faults of contact printers now assume much greater relative im- 

The modulated oscillator test described by Baker 5 has provided 
a very sensitive tool for checking performance, and has facilitated 
making comparisons capable of showing relatively small differences. 
Thus the advantages of the non-slip printer can now be demonstrated 
in comparison with even the best conditions attainable on a sprocket 
type of printer. 

As has been explained in previous publications, the mean linear 
velocity of the raw stock as it passes the printing point is determined 
by the direction from which it approaches the point. Thus, as the 
guide-roller shown in Fig. 1 is at position A , the raw stock approaches 



[J. S. M. P. E. 

FIG. 1. Drum and guide-roller arrangement in general 
use in non-slip printers. 

FIG. 2. Printer with sprocket so located as 
to reduce loop-length variations, and vector 
diagrams showing film tension required to dis- 
place guide-roller. 


the printing point P from the right, and therefore presents a concave 
or compressed surface against the negative that drives it. This is 
the condition when the raw stock is long compared with the negative. 
If the films are of substantially equal length, the loop becomes shorter 
and the guide-roller is drawn over to position B so that the raw stock 
presents a convex surface where it is pressed against the negative. 

However the operator may thread up the machine, the guide- 
roller reaches a stable position within a few feet of film, and thereafter 
there is a fixed relation between the place where a given sound ap- 
pears on the negative and where it appears on the print. Obviously 
the operator can not control this relation any more closely than he 
can anticipate the final running position. If, for example, he threads 
the printer with such relation between the synchronizing marks on 
the sound and picture negatives as to obtain perfect synchronism 
with the roller in the middle position, and the film condition happens 
to be such that the roller moves to the right, the sound will be printed 
slightly ahead of the ideal position, on account of the unanticipated 
increase in loop length. Deviations from exact synchronism from 
this cause have been well within tolerances, but inasmuch as there 
are numerous other causes of lack of synchronism and the effects may 
be cumulative, it is desirable to minimize even this small variation. 

Fig. 2 shows an arrangement adopted in one of our models of 
non-slip printer, in which the variation is less owing to the oblique 
angle at which the sprocket is placed. This arrangement however, 
has a disadvantage in that the film tension required to hold the guide- 
roller near the left-hand limit of its swing tends to become excessive. 
Adjacent to the roller, shown in the two extreme positions, are force 
vector diagrams showing the tension on the film required to exert a 
given force upon the roller in the direction in which the roller is free 
to move. It is practically necessary to apply, by means of spring or 
weight, a biasing force on the roller in order that it may control the 
shape of the loop. It is this biasing force that must be overcome by 
the film tension. The expedient of placing the sprocket in an ob- 
lique position, as shown in Fig. 2, is not an altogether satisfactory 
solution of the problem of reducing the loop-length changes. With 
the guide-roller near the right-hand position, the action differs only 
slightly from that of the arrangement shown in Fig. 1. Therefore, 
a large part of the length variation is still required for covering this 
part of the range. On the other hand, near the left-hand position 
the required film tension rises almost abruptly to values that may 

140 E. W. KELLOGG [J. s. M. P. E. 

cause danger of slipping. For this reason, we have recommended 
designs more nearly in accordance with Fig. 1. 

Any loop arrangement in which loop shortening is accompanied 
by increase of tension has an insidious quality of affording a slipping 
printer, which in appearance functions like a non-slip printer. 

When film is held in contact with a moving surface (either a drum 
or the surface of another film) and is subjected to too much tension 
for true non-slip operation, the slipping seems to be almost of the na- 
ture of a creep. It is too gradual and continuous to be noticeable to 
the eye in the appearance of the film loop, but sufficient in magnitude 
and irregularity seriously to impair the sound reproduction or the 
printing. Models of what were intended to be non-slip printers have 
been built in which it turned out that the range of possible 
compensation due to curvature control was inadequate, but when 
the raw-stock loop was short it was also under considerably increased 
tension and the machine had the appearance of functioning 
perfectly. Only careful tests showed that with the short raw-stock 
loop the apparent compensation was achieved in part by curvature 
and in part by slipping. This danger is not necessarily confined to 
machines in which the geometrical relations of Fig. 2 are employed. 
If the biasing spring in a design along the lines of Fig. 1 is short or too 
stiff, the tension required to hold the guide-roller in position B will be 
much greater than the film tension for position A . The ideal biasing 
means would be a counterweight, but a properly designed spring will 
approximate the action of a counterweight. It is necessary only that 
the average spring stretch be large compared with the change in 
stretch or extension, between positions A and B . As further assurance 
against slipping, it is advisable to maintain the tension at all times 
as low as practicable. This means that the guide-roller must be 
pivoted with the minimum of friction. 

Assuming that the guide-roller is so mounted that there is little 
friction to be overcome and that the biasing force is kept low, there is 
still a certain amount of mechanical work that must be supplied to pull 
the roller from the long to the short-loop position. The mechanical 
work that the film supplies is the product of tension multiplied by the 
change of loop length. We must therefore permit some change of 
loop length, although it is feasible to make the change considerably 
less than is required with a design such as shown in Fig. 1. In order 
to reduce the variations of loop-length, an arrangement analogous to 
a lever must be employed to step up the motion, at the cost of re- 




FIG. 3. Double-roller arrangement in which loop length 
is independent of guide-roller position. 

FIG. 4. Double-roller system similar to Fig. 3. in 

FIG. 5. Modification of Fig. 4, such that film 
tension will cause movement of rollers. 

142 E. W. KELLOGG [J. s. M. P. E. 

quiring greater film tension. In general, this means the employ- 
ment of a second roller connected to the guide-roller in such a way 
as to offset the effects of the guide-roller; or, in other words, if the 
guide-roller moves in a direction that would tend to shorten the loop, 
the second roller moves in a direction that would tend to lengthen the 
loop. The required relation may be most easily appreciated by con- 
sidering an arrangement of rollers by which the movement of the 
second roller completely cancels the effect of movement of the guide- 
roller, making the total loop length substantially independent of the 
position of the rollers. Such an arrangement is shown in Fig. 3. 
This, of course, is not a usable arrangement unless some other means 
of controlling the roller positions is provided. In Fig. 3, the guide- 
roller and auxiliary roller are maintained at constant separation by 
a link L. The arm Ai, on which the guide-roller is supported, is 
pivoted at such a point that the arm and the stretch of free film FI 
are substantially parallel and of equal length. Likewise, the arm 
A 2, which supports the auxiliary roller, is parallel to the stretch of 
film F 2 between the sprocket and auxiliary roller. Except for very 
slight shortening of the loop in the extreme positions where the film 
follows an appreciable arc around the drum, or sprocket, or the ad- 
jacent rollers, the three stretches of unsupported film in Fig. 3 are of 
constant length, and the sum of the angles of wrap around the two 
rollers is likewise constant, making the total loop length constant. 
So long as the guide and auxiliary rollers are a constant distance 
apart, the film between them will be of constant length, whether it 
passes around them in the same direction or in reverse directions, as 
shown in Fig. 4. In Fig. 4 we have the sprocket in more nearly the 
position that has proved to be convenient from the design standpoint. 
The same reasoning that was used to show that the film loop is of 
constant length in Fig. 3 may be applied to Fig. 4. It is now very 
easy to see how the linkage may be modified so that the auxiliary rol- 
ler will move slightly less than the guide-roller. Such a change can 
be effected by shifting the link downward on both arms as shown in 
Fig. 5. We now have an arrangement in which tension of the film 
will cause the rollers to move toward the left, and the links may be so 
proportioned that the total change of loop length is as small as de- 
sired, the limit being determined by permissible film tension. In 
this it must be remembered that the smaller the change of loop length 
the greater will be the required film tension in comparison with the 
biasing force on the roller. It is thus especially important when em- 



ploying a linkage of this type, to keep the biasing force as small as 

One of the objections to the arrangements shown in Fig. 2, and 
to a less degree than that shown in Fig. 1, is that the wrap around the 
guide-roller is none too large for desirable guiding conditions when 
the guide-roller is in the right-hand position, and the wrap becomes 
less when the roller reaches the left-hand position. It will be noticed 
that in Fig. 5 a large angle of wrap around both rollers is maintained 
at all times, while the reverse bend in the film is of assistance in forc- 
ing the film between the flanges of the guide-roller. This condition 
makes it possible to achieve satisfactory guiding with less film tension 
than would be required for smaller angles of wrap. This in turn al- 

FIG. 6. Alternative method of providing control by 
film tension. 

lows the employment of a very light biasing spring or counterweight. 

Instead of modifying Fig. 4 by altering the position of the con- 
necting link, a similar effect may be obtained by shifting the sprocket 
and pad-roller slightly to the left, as shown in Fig. 6, so that the 
stretch of film Fz and the arm Az are no longer parallel. The amount 
of this shifting of the sprocket controls the shortening of a loop. 

It may not prove convenient to locate the centers of the swinging 
arms as shown in Figs. 5 and 6. There are numerous variations that 
will give substantially equivalent results. Fig. 7, for example, shows 
a linkage in which the shorter arm is adjacent to the longer stretch of 
free film, and the longer arm adjacent to the shorter stretch. As 
soon as we depart from the parallelogram arrangement as shown in 
Figs. 3 to 6, it becomes necessary to study very carefully the rate of 



[J. S. M. P. E. 

change of loop length with respect to roller displacement. The ideal 
is that this should be constant, each increment of roller movement 
being accompanied by a proportionate change of loop length. This 
relation will assure practically constant film tension throughout the 
range of positions. It was pointed out in discussing Fig. 2 that the 
ability of the film to produce adequate force in the direction of roller 
movement became poor for certain positions. With an arrangement 
such as shown in Fig. 7, the force exerted upon the guide-roller by 
the film becomes less as the roller moves to the left, and, in particular, 

FIG. 7. Modification of Fig. 4 to meet exigencies 
of design but having essentially same character- 

the horizontal component of the resultant force decreases rapidly. 
At the same time the opposing horizontal force exerted by the film 
upon the auxiliary roller tends to increase. The effect is compen- 
sated by the large change of angle of the short arm At. The radial 
thrust of this arm has a component in the right direction to relieve 
the film. At the same time the arm A\ and the link L fall more 
nearly into line, thus affording a toggle effect enabling a given force 
exerted upon the guide-roller by the film to overcome a larger force 
exerted by the film upon the auxiliary roller. The design of such a 
linkage is largely a matter of cut and try. If it is not well done, 
there are likely to be hard spots where the film appears to work al- 



most against a dead center, while at some other part of the swing 
there will be unnecessarily large changes of loop length. The analy- 
sis may be made by calculation, or graphically, or by trial with a 
model. Fig. 8 shows a model used for this purpose. The lengths 
of arms and locations of pivots and rollers could be changed quickly, 
and the action could be checked with reasonable accuracy by pulling 
the film and observing the movements. 

In case design limitations make it impossible to obtain the desired 
relation between loop shortening and roller movement, and it is neces- 
sary to resort to some auxiliary device to keep the film tension more 

FIG. 8. Model used for determining suitable 
linkage proportions. 

nearly constant, a toggle spring or a counterweight above a pivot 
point, as shown in Fig. 9, might be employed. It is far preferable, 
however, to work out linkage proportions that will make this un- 
necessary, for if the film must for any position work too near a dead 
center, no amount of reduction in the force exerted by the spring will 
enable it to function properly. Fig. 10 shows the roller and link 
arrangement employed in a laboratory model of non-slip printer. 
This linkage was worked out with the help of the model shown in 
Fig. 8. H. A. Backus, who has cooperated in this development, 
analyzed a number of arrangements, both graphically and by calcu- 

146 E. W. KELLOGG [J. S. M. P. E. 

lation, before arriving at the proportions adopted, and also checked 
the contemplated design experimentally. 

One of the requisites for avoiding too much variation in loop 
length is to keep the guide-roller as close to the pressure-roller as 
possible. The employment of a small-diameter guide-roller also 
assists in bringing about this result, but the small-diameter guide- 
roller is undesirable in that it affords less satisfactory guiding, and 
more tension is required to make the film ride snugly against the 
guide-roller. If a compensating linkage, such as has been discussed, 
is employed, it becomes less necessary to keep down the size of the 
guide-roller. In fact, the total required variation in loop length is 
not materially altered by adoption of a fairly large diameter guide- 
roller, provided the linkage is de- 
signed to suit the roller size chosen. 
If the stretches of unsupported 
film took the form of straight tan- 
gents from one roller to the next, 
there would be no purpose in 
complicating the design beyond 
the adoption of such a linkage 
as has been described. There is, 

however, a loss in range of angle 

FIG. 9. Toggle spring and ' . 6 . 

overhead counterweight used to due to bowing, as shown in Fig. 11, 
equalize film tension in certain w hich must be offset by increased 

movement of the guide-roller. This 

bowing of the film becomes greater when the tension is kept low 
but is somewhat helped by the employment of a larger guiding 
roller. Fig. 12 shows the manner in which an extra movable roller 
may be employed further to minimize the changes of loop length. 
The extra roller is not needed and does not come into play 
when the guide-roller is in the right-hand position, or, in other 
words, when the film is bent around the drum. In fact, there is 
no room for such a roller under these conditions. It comes into 
play as the guide-roller swings to the left, and presses against 
the free film sufficiently to produce a slight reversed bend, thereby 
causing the film to follow the pressure-roller more closely and bring 
about the desired change of angle. If the linkage of the type shown 
in Fig. 12 is such as to maintain the two rollers at constant separa- 
tion, some special arrangement would be required for threading the 
film. If, however, the linkage is so designed that the extra roller 



FIG. 10. Non-slip printer layout employed in ex- 
perimental model. 

FIG. 11. Loss of control of angle 
through bending of film. 

FIG. 12. Employment of auxiliary 
roller to increase range of angle for 
given movement of guide-roller. 

148 E. W. KELLOGG [j. s. M. p. E. 

has slightly more movement than the guide-roller, the separation 
can be caused to increase sufficiently with movement toward the right, 
so that threading may be accomplished by merely swinging the roll- 
ers to extreme right-hand position. 

Although the method of driving the drum in a non-slip printer 
bears no necessary relation to the general subject of this paper, it is 
appropriate in any discussion of such printers to give consideration 
to the possibility of overdriving the drum. From the standpoint 
of guiding the negative there is an advantage in supplying enough 
torque to the drum, by means of an auxiliary drive such as the mag- 
netic drive used in recorders, to cause the negative to run moderately 
tightly where it is fed onto the drum, and loosely where it leaves. 
The auxiliary drive also helps to bring the drum up to speed. Cau- 
tion must be observed to insure that the torque supplied is not suf- 
ficient to cause slipping of the negative on the drum. A soft- tired 
pressure-roller below the drum, or on the side of the slack loop, is 
usually employed when the drum is overdriven. Such a pressure- 
roller provides assurance against slipping, but is not entirely neces- 
sary in the printer since the main pressure-roller insures a fair amount 
of traction. If the extra pressure-roller is used, precautions need to 
be taken to prevent the film from running off during starting. The 
film stability problem has been discussed in detail in an earlier paper 
on recorders. 6 

Acknowledgment should be made of the valuable assistance of 
H. A. Backus of the Engineering Department of RCA Manufactur- 
ing Co. 


1 BEDFORD, A. V.: U. S. Patent No. 1,754,187. 

* BEDFORD, A. V.: U. S. Patent No. 2,098,371. 

* BATSEL, C. N.: "A Non-Slip Sound Printer," /. Soc. Mot.Pict. Eng., XXIII 
(Aug., 1934), No. 2, p. 100. 

4 WOOD, R. V.: "A Shrinkage- Compensating Sound Printer," /. Soc. Mot. 
Pict. Eng., XVIII (June, 1932), No. 6, p. 788. 

6 BAKER, J. O., AND ROBINSON, D. H.: "Modulated High-Frequency Record- 
ing as a Means of Determining Conditions for Optimal Processing," /. Soc. 
Mot. Pict. Eng., XXX (Jan., 1938), No. 1, p. 3. 

6 KELLOGG, E. W.: "A New Recorder for Variable-Area Recording," /. Soc. 
Mot. Pict. Eng., XV (Nov., 1930), No. 5, p. 653. (The title of this paper is mis- 
leading, in that the main features of the recorder were in no wise related to the 
type of sound-track to be recorded.) 



MR. ROBERTS: In that type of printer it has been our experience to find more 
trouble in controlling the raw stock than the negative. Are there any particular 
advantages to positive stock guidance in overdriving the drum? 

MR. KELLOGG: The overdrive has nothing to do with the action of the raw 
stock, but the suggested guide-roller arrangement will, I believe, help the guiding. 

MR. CRABTREE: Are there any advantages in increasing the weight beyond 
the slippage point? 

MR. KELLOGG: I believe not. We usually like to keep it as light as we can. 
I should be very much interested to know what experience the Consolidated 
Laboratories have had in regard to static. That has at times become a serious 
problem, and it is aggravated by excessive pressure. I understood that by less- 
ening the pressure, they were able to alleviate it. Their first tests were run at 
180 feet a minute. That, plus considerable pressure, made static markings on 
the film. What was finally done to cure the trouble, I do not know. 

MR. DAVEE: Those who have made comparisons of the non-slip printer 
with other methods of printing, I believe, can appreciate the advantage of print- 
ing sound with this printer. Would we not be able to obtain much better picture 
quality, comparable probably with the step printer, in contrast to the continuous 
printer, and thereby improve the picture quality considerably? 

MR. KELLOGG: The main problem is, of course, that of framing. A picture 
printer must operate so that if the negative is made with a picture frame line 
directly opposite a pair of sprocket holes, it will make a print of which the same 
is true. 

The non-slip printer for sound is subject to slight variation or uncertainty in 
the position of the print with respect to the sprocket holes. It is only at the 
sprockets that we have any control of the relative positions of the two films as 
they run through the machine, whereas the printing contact is on the drum with 
the variable loop between. Therefore, we can not make a continuous picture 
printer along these lines, and be sure that the pictures will stay in frame. I 
believe that a non-slip picture printer is a possibility, but I do not see how to 
utilize the Bedford sound printer principle for that purpose. 

MR. ROTHBERG: Have you found any practical answer with regard to the 
extremes of diameter of the sound drums? 

MR. KELLOGG: Such experience as I can report seems a little contradictory. 
We have furnished to those who requested them, prints of a design embodying 
our ideas at the time the drawings were made. These drawings showed a drum 
about an inch in diameter, and a pressure-roller about 5 /s inch in diameter. In 
our laboratory work these diameters appeared to take care of all the variations 
of negative shrinkage that we had encountered. In our Canadian laboratory 
we never seemed to find the guide-roller running anywhere but pretty well toward 
the left, which would indicate that a larger diameter drum could be used. If 
the drum were made a little larger, the guide-roller would not have to run so far 
to the left, because the negative surface would not be stretched quite so much. 
Our laboratory model saw considerable service in Hollywood, and there I believe 
Mr. Blaney found it desirable to use a 7 /s-inch drum. After the machine was 
returned to Camden, experience seemed to indicate that we did not need quite 
so small a drum, and we went back to a 1-inch diameter. 



Summary. The only requirements of a buzz-track are that the track be of correct 
width and located properly with respect to the edge of the film nearest the track, and 
that the sound produced by a weave in one direction shall be readily distinguishable 
from the sound that results when the film is displaced in the other direction. 

It is desirable that the buzz-track film should be a direct recording rather than a 
print, since there is then less chance of inaccurate location. A simple recorder has 
been constructed for the sole purpose of making 35-mm. buzz-track film. It can read- 
ily be converted for 16-mm. film. All possible precautions are taken to insure cor- 
rect track width and location. In view of the small quantity of buzz-track required, it 
is contemplated that only one such machine will be needed. 

4 'Buzz- track" is the name given to a special recording made for the 
purpose of adjusting reproducers so that the scanning beam will fall 
at the correct location with respect to the guided edge of the film. 
Fig. 1 is an enlargement of a small section of 35-mm. buzz- track. 
The space or strip normally scanned by a correctly located repro- 
ducing light-beam is free from modulation, but on either side of this 
strip is recorded a tone. A one-thousand-cps. note has been recorded 
on the side next the sprocket holes and a 300-cps. note on the side 
farthest from the sprocket holes. If the film-guiding and optical 
systems of the projector under test are adjusted in proper relation 
to each other, no tone is produced. If the projector needs adjusting, 
a tone is heard, the frequency of the tone telling the service man 
which way to shift the film-guiding roller. On 35-mm. buzz-track 
film the modulation-free strip should be centered 0.2435 inch from 
the edge of the film, and should be 0.088 inch wide, which would 
clear a correct scanning beam by 0.002 inch on each end. The cor- 
responding figures for 16-mm. buzz-track are 0.058 inch from edge 
of the film to the center of the track and 0.069 inch clear width. 

Up to the present time, buzz-track films have been made by first 
recording a special negative and then making as many prints of the 

* Presented at the Fall, 1937, Meeting at New York, N.Y. ; received October 
11, 1937. 

** RCA Manufacturing Co., Camden, N. J. 



negative as are required. It should not often be necessary to make a 
new buzz- track negative, but when it is necessary considerable labor 
is involved. One method is to shift the recording optical system in 
a standard recorder some twenty or thirty, mils to the side, place a 
mask at the slit and by trial locate the mask so that one edge of the 
desired clear space falls at exactly the right distance from the edge 
of the film. The required footage of negative film is then run through 
the recorder. The optical system is now shifted toward the other 
side, the mask readjusted and the correct position determined by 
successive trials, each trial requiring processing a piece of test-film. 
The galvanometer is then modulated at the desired frequency, and 
the negative run through the recorder again, to record the other 

FIG. 1. Sample of buzz- track. 

side of the buzz- track. In addition to the numerous trials required 
to adjust the optical system correctly, making a buzz-track negative 
means taking the recorder out of service and upsetting the existing 
adjustments. Making a negative and then printing it is furthermore 
undesirable in that possible imperfections in the guiding of the printer 
will cause errors in the location of the printed track and there will 
also be variations due to the combined effects of shrinkage of nega- 
tive and print. 

It recently became necessary to make an additional buzz-track 
negative, and it appeared that a special recorder might be built that 
would cost little more than the labor required for setting up the special 
recording system and making a negative in the manner just de- 
scribed. Once such a recorder is built, it would make further pro- 
duction of buzz- track film a very simple matter. Although there 

152 E. W. KELLOGG [J. S. M. P. E. 

is no reason to suppose that additional machines will be built, a 
brief description of the design may be of interest. 

In only one respect is accuracy required in a buzz-track. The 
modulation-free strip must be of correct width and accurately located 
with respect to the edge of the film. For once "wows" may be for- 
gotten. The tones recorded on the two sides of the track must be 
readily distinguishable, but there is no other requirement as to the 
choice of tone. Fig. 2 shows a front view and Fig. 3 a rear view of 
the machine. The optical arrangements are shown in Fig. 4. 

The film drum is driven by means of a string belt at approxi- 
mately 90 feet per minute. A three- watt lamp is located at F in 
Fig. 4. The lamp filament is a small-diameter helix, and is imaged 

FIG. 2. Buzz-track recorder, front view. 

upon the film by an objective lens LI. The reduction ratio is such 
as to give an image height of the order of three mils. In the hori- 
zontal plane, the cylindrical lens LI throws the filament out of focus 
and changes the image on the film into a horizontal line of sufficient 
length to record the wide track that is needed. Lenses LI and L 2 in 
combination focus the mask M upon the film. Since Z/2 is a cylinder, 
only vertical edges in the mask are imaged upon the film. An aper- 
ture in the mask is divided into two openings by a vertical bar in- 
dicated at BI. The result of the combination is a rather long line 
of light upon the film, with a break in the middle. The shadow of 
the bar upon the film determines the position and width of the modu- 
lation-free portion of the track. The mask was set up without the 
tone wheel W, and the sides of the bar BI were filed until their 

Feb., 1938] 



images fell upon the film at exactly the right positions. The tone 
wheel W has ten notches in the periphery, and just inside the notches 
is a row of thirty- three holes. The opening to the right of BI is un- 
covered ten times per revolution of the tone-wheel, while that to the 
left is uncovered thirty- three times. The speed of the motor is such 
that this results in tones of approximately 300 and 1000 cps. The 

FIG. 3. 

Rear view of buzz-track recorder, showing motor, tone- 
wheel, and reversible mask. 

exposure changes rather suddenly from full intensity to zero as the 
wheel is revolved, giving for each tone a series of black spots and 
clear spaces. 

It was desirable that the same machine should be capable of making 
either 35-mm. or 16-mm. buzz-track. In order to make the change- 
over simple, the mask is mounted on closely fitting studs, and another 
pair of windows is cut in the opposite side. Simply reversing the 
mask produces an image on the film having the correct separation 
between recorded tones for 16-mm. buzz- track. The position of the 
recording beam with respect to the edge of the film must also be cor- 



[J. S. M. P. E. 

rect. Owing to the fixed position of the rows of holes in the tone- 
wheel, this shift in the track location could not well be taken care 
of at the mask. Therefore, the machine has been provided with 
interchangeable drums having slightly different flange locations. 
When making 35-mm. buzz-track, the drum is driven directly 
from the motor by a string belt. When 16-mm. buzz-track is being 
made, a pair of intermediate pulleys is employed so that the film is 
propelled at about normal speed, thereby giving approximately the 
same two tones as used for the 35-mm. buzz-track. 

FIG. 4. Main elements of buzz-track recorder. 

Since the machine will be used only at long intervals, no provision 
has been made for supply and take-up reels. When a few hundred 
feet of buzz-track are to be recorded, a film rewind may be placed 
near the machine and an assistant operate the rewind. Supply and 
take-up reels could, of course, be added easily if warranted. 

As was pointed out at the beginning of this paper, there is one 
respect in which precision is required. There can be little doubt 
that the most accurate lateral guiding may be attained by providing 
the drum with a flange and insuring that the film will be held snugly 
against the flange. A stationary guide feeds the film to the drum 
from the proper position. As the film is fed to the drum it passes 


under a rubber-tired pressure roller which is mounted on a skew 
axis. This roller exerts a continuous thrust upon the film, tending 
to hold it against the flange. As the film passes the recording beam, 
it is held against the drum by a belt which passes around two idler 
pulleys and over the drum. The belt is under sufficient tension to 
exert the required pressure between the film and the drum! The 
idler pulleys are offset in such a manner that the belt must slide upon 
the surface of the film, exerting an additional axial force tending to 
hold the film against the flange. End-play in the drum shaft must 
be practically eliminated. It appears that this condition may be 
satisfactorily approximated by closely adjusting the pulley on the 
shaft. A spring washer may easily be provided if found desirable, 
in order to insure that the thrust surface on the drum will always 
run against the end of the bushing. 

Recording buzz-track on this machine involves very little more 
labor than to run the same amount of film through a printer. 
There is therefore little to be said for resorting to the negative-posi- 
tive method of making buzz-track films, whereas there is a distinct 
disadvantage in such a method because of the chance of wrongly 
locating the track. The directly recorded buzz- track, of which 
Fig. 1 is a sample, will differ from the buzz-track previously employed 
in that the modulation-free strip is clear instead of black. This 
may result in slightly more ground-noise, but unless the film has been 
abused, the ground-noise will always be so low that it will not mask 
either of the recorded tones, even though the light-beam extend into 
the recorded area ever so slightly. 


Summary. A discussion of some of the practical aspects of push-pull recording 
supplementing a discussion of the theoretical principles by 0. 0. Ceccarini on page 
162 of this issue of the Journal. 

Sound that is to accompany motion pictures must necessarily be 
reproduced with the greatest possible accuracy in order to create the 
greatest illusion of reality. In seeing motion pictures the audience 
is aroused to a sound expectancy that is not necessarily essential to 
the radio or the phonograph. It is necessary that the recording and 
reproducing systems have frequency and volume ranges commen- 
surate with those of the original sounds, and a high degree of linearity. 
To approximate this condition, a frequency range of 50 to 8000 cps. 
is necessary, and a volume range of 50 to 60 decibels. Reproducing 
systems are now available, and are being installed in theaters, that 
will meet these requirements, and recording practice has been de- 
veloped to such a degree that it also fulfills the requirements. How- 
ever, it has been found necessary to limit the volume range of releases 
until such time as the majority of theaters are adequately equipped. 

To date, standard methods of recording on film, both variable- 
density and variable-width, have given a volume range of approxi- 
mately 40 decibels, at the expense of considerable distortion. In 
the variable-density system the principal limitation has been the 
small linear range of density between the toe and the shoulder of 
the characteristic curve of positive stock, plus the distortion due to 
the noise-reduction system because of its slow operating time. Adop- 
tion of the push-pull feature would reduce these limitations by cancel- 
ling these internal distortions and, accordingly, development of such 
a project was undertaken. 

The light-valve as shown in Fig. 1 has been modified to have two 
pairs of biplanar ribbons of the conventional type, each pair exposing 

* Received October 11, 1937. The theoretical principles of push-pull re- 
cording are outlined by O. O. Ceccarini, p. 162 of this issue. 
** Metro-Gold wyn-Mayer Corp., Culver City, Calif. 



half of the sound-track. The valve is so connected that the ribbons 
act in the push-pull manner. Mechanically it is impossible to mount 
the ribbons in line with each other. In order to scan the track with a 
single slit, the image on the film is optically relocated. This is ac- 
complished by placing an optical flat in the path of each light-beam 
in the form of a saw-buck as shown in Fig. 2. This arrangement 
moves the axis of the path in proportion to the angle at which the 
optical flat is interposed. As a result, the two lines of light formed by 
the valves are colinear on the film. 

Any power that is added to the signal component to effect noise- 
reduction appears in phase on the track and is cancelled out in the 
push-pull reproduction. Since these added signals appear on the track 
in the form of changes of density, then, if the density is the same on 

FIG. 1. Push-pull recording light- valve. 

either side, zero signal results on the external side. Now it is possible 
to speed up the noise-reduction circuit so it will operate faster. This 
higher speed decreases the amount of clash, since the light- valve will 
clash at lower amplitudes if the noise-reduction is not able to pull it 
out of the way fast enough. The push-pull method allows greater ex- 
posure of the film, and in order to achieve maximum cancellation it 
is necessary to keep the exposure uniform from one side to the other. 
This is accomplished by checking the density gradient from time to 
time, and by adjusting the flat-ribbon filament lamp so that its field 
is as uniform as possible. Very often it becomes necessary to select 
lamps carefully when wide variation in manufacture is encountered. 
If a recording is made in which the two halves of the tracks are re- 
corded in phase, then when the recording is reproduced on a balanced 
machine the output will be a measure of the cancellation. 

In the variable-width class A push-pull system, 1 the images of the 
triangular slits are so placed that they are symmetrical with respect 



[J. S. M. P. E. 



FIG. 2. Optical arrangement for push-pull recording light-valve. 










FIG. 3. Light train of RCA MI-1070 reproducer. 

Feb., 1938] 



to the center of the slits. Then the center of the track is made to 
correspond with the center of the slits. The residual width of the 
track will then give a measure of balance. 

Fig. 3 shows schematically an RCA MI-1070 sound-head, consisting 
of a push-pull photocell, a special lens, and a prism assembly, to- 
gether with a push-pull photocell transformer. This combination 
makes it possible to reproduce either single-track or push-pull record- 

The 920 photocell contains two anodes and two cathodes. When 
connected through the selector switch for single-track reproduction, 


u ; 


RCA 920 

J s : 

> TO 








-RCA 920 











FIG. 4. (Upper) Push-pull circuit; (lower) standard circuit. 

the cathodes are connected in parallel. The photocell then operates 
in the same manner as the standard cell. For push-pull reproduction 
the two cathodes are separated, and operate alternately through the 
photocell transformer (Fig. 4). 

To balance the equipment for push-pull recording, any bilateral or 
standard variable-density film can be used. When set for push-pull 
reproduction, the balancing potentiometer is adjusted for minimum 
reproduction of the recording sound at high volume. Another method 
that is currently used in studio review rooms is to use alternating cur- 
rent as the source of supply for the exciter lamp and adjust the 



[J. S. M. P. E. 

balancing potentiometer for minimum output. This type of sound- 
head is known as a direct scanning reproducer. 

Fig. 5 shows another type of sound-head, manufactured by the 
Western Electric Company. 2 It employs a film scanning system 
known as the "projection, rear, or indirect type," which has been 
used in re-recording reproducers and consists essentially of an exciter 
lamp, condenser lens-prism assembly and objective lens, and a scan- 
ning slit, behind which is a collimating lens and a photocell. 

The condenser lens-prism assembly should be adjusted so that it 
focuses the filament image some distance in front of the film plane 
on the lamp side so that the film is illuminated with a blob of light. 

FIG. 5. Western Electric "projection" type sound-head. 

The objective lens is adjusted to focus the track image sharply upon 
the scanning slit, whose width is approximately 12 mils. The drum 
that holds the film in place at the point of scanning is of the rotary 
stabilizer type, similar to that in the RCA sound-head. Fig. 6 shows 
the wiring diagram of the TA-7400 reproducer set for both push-pull 
or single track recording. 

To obtain the maximum benefit from the increased volume range, 
it is necessary that flutter due to the motion of the film at the point 
of scanning be reduced to a minimum. Two types of flutter are 
encountered. One is 96-cycle flutter (sprocket-hole frequency) 

Feb., 1938] 



which results from the fact that when sprocket-holes are used to pull 
the film the motion is not continuous. For that reason, methods that 
use sprocket-teeth have been eliminated in favor of drum scanners. 
The drum also reduces low-frequency flutter when the rotary stabi- 
lizer 3 principle is applied. 

Push-pull variable-density recording has been in commercial use 
since the early months of 1935, and at the present time both the 
variable-density and variable-width methods are in wide use among 
the studios for original recording. The release of movietone push- 
pull sound-track is very limited, due to the small number of theaters 

6-A P.E.C. / 



C c 










50,000 A 



200 V. 

FIG. 6. Western Electric System TA-7400 reproducer set. 

equipped with push-pull reproducers. However, during the past 
year several pictures have been released with a limited number of 
push-pull copies. It is expected that within a short time enough 
theaters will be capable of playing push-pull so that it will be prac- 
ticable to release push-pull prints on a larger scale. 


1 DIMMICK, G. L.: "RCA Recording System," /. Soc. Mot. Pict. Eng., XXIX 
(Sept., 1937), No. 3, p. 258. 

2 DAVIDSON, J. C.: "A New High-Quality Film Reproducer," /. Soc. Mot. 
Pict. Eng., XXVIII (Feb., 1937), No. 2, p. 202. 

a COOK, E. D.: "Technical Aspects of the High-Fidelity Reproducer," XXV 
(Oct., 1935), No. 4, p. 289. 



Summary. A discussion of the mathematical principles underlying the push-pull 
method of sound recording, previously discussed by Douglas Shearer at a joint meet- 
ing of the Pacific Coast Section of the SMPE and the Academy of Motion Picture Arts 
& Sciences on June 4, 1936. 

The push-pull system of recording sound was first brought out for 
the purpose of improving the method of reducing background noise. 
It might be well, therefore, to state the principles upon which the 
system is based. 

Assume a single frequency applied to a double light-valve or modu- 
lator, and let the customary sound-track be divided exactly into two. 
Record the positive half -cycle on one track and the negative half -cycle 
on the other, each half -cycle correctly displaced upon the film. For 
convenience of analysis assume these records to be made in the vari- 
able-width fashion. 

Each track has then a sound record identical to the output of a 
half -wave rectifier of ideal characteristic (Figs. 1 and 2) . 

As a common starting point we may represent the track containing 
the positive half -cycles by /i (x) and the track representing the nega- 
tive half -cycles by/ 2 (x). Both/i (x) and/ 2 (x) can be expanded in 
terms of a Fourier series, the necessary requirement being that the 
two series must be capable of representing respectively : 


/!(*) = E sin * < X< ir 

= * < x < 2* (1) 

< x < -w 

E sin x TT < x < 2ir (2) 

where E represents the maximum amplitude. 

* Received October 6, 1937. 
** Consulting Engineer, Metro-Goldwyn-Mayer Corp., Culver City, Calif. 



These series are readily found to be 

/,(*) =:?+? sin x + \] cos nx ( n even) W 

TT 2 7r(l n-) *-^ 


7? Tf O 77 \ % 

(x) = __-(-_ sin # / cos nx (n even) (4) 

TT 2 7r(l - W 2 ) ^ 

The algebraic sum of fi (x) and / 2 (#) is obviously 

/i() + MX) = E sin x (5) 

This algebraic sum is readily performed by connecting two photoelec- 
tric cells in push-pull position, as is done with vacuum tubes, from 
which type of connection the name was derived. 

It is obvious by inspection that the amount of noise-reduction is 
the greatest possible without any additional equipment for the pur- 
pose. The ratio of signal amplitude to background noise is maximum 
and remains always constant. 

The critical requirements of the system are that one record must 
not only be an exact counterpart of the other, but, also, that the line 
of division of the record must be exactly along the neutral axis. In 
projection the illumination of both tracks must be identical, and every 
electrical part up to the point of combining the records must be per- 
fectly symmetrical. 

It is physically impossible to close the ribbons of the light-valve 
completely to permit recording in the above-described manner. 
More likely the two records would be of the following forms (Figs. 3 
and 4) : 

fi(x) = E sin x -a<x<Tr + a 

= TT + a. < x < 2ir - a (6) 

f 2 ( x ) =0 < x < IT a 

= E sin x Tr-a<x<27r + a (7) 

where a is an appreciable angle. For this type of record the former 
discussion no longer applies. 

Following the method adopted for the classification of amplifiers 
we have found it convenient to assign the term class A push-pull 
record to a record in which the wave is fully or integrally recorded 
in each track, and class B push-pull record to a record in which one- 
half of the wave is recorded on the first track and the following half 
on the second track. 



[J. S. M. P. E. 

FIG. 1. 
fi(x) = E sin x O<X<TT 

= Tr<X<2ir 

\J \J 

FIG. 2. 
MX) = 0<x<ir 

o< <x 


FIG. 3. 

= sin x 

= 7r+a<JC<2ir-a 

O tr 

FIG. 4. 

= E sin 



In accordance with this classification, then, the type of record repre- 
sented analytically by conditions 1 and 2 is a class B record, while 
the record specified by conditions 6 and 7 is a class A-B record. 

The following important considerations are self evident: In re- 
producing class A-B records, the class A components will add numeri- 
cally while the class B components will add algebraically, with the 
result that extremely weak sounds consisting only of class A com- 
ponents will appear twice as loud in proportion to stronger sounds. 

With the variable-density system, correction for this irregularity 
can not be made. In the case of the variable- width system it is con- 
ceivable to subtract the class A components physically by suitable 
masks during reproduction, thus reducing the record to pure class B. 
The critical requirements of precise balance of the output circuits, 
faultless alignment and azimuth adjustment of the reproducing slit 
with respect to both tracks, and perfect balance of light on both tracks 
are, however, common to both systems. 

It might, perhaps, be worth while to point out that poor alignment 
and azimuth adjustment of the slit will fail to produce perfect cancella- 
tion of the term 

2E v^ 

- - > cos nx 

only at high frequencies, while unbalance of the circuits and un- 
balance of light will have identical effect throughout the whole fre- 
quency range. 

Before entering into details about the class A push-pull method 
it will be convenient to digress slightly to the subject of noise-reduc- 

The pulsating direct current used to bias the light- valve is usually 
obtained by rectifying the audio signal by means of a full- wave recti- 
fier. The open-circuit voltage output of this device can also be repre- 
sented in terms of a Fourier series of the following form : 

"* nx (n cven) 

The term 2 A fir is the direct-current term used to bias the light- valve, 
while the harmonic sum represented by the second term at the right 
is usually more or less suppressed by an electrical filter. Because of 
this filter the direct-current term does not build up as fast as the 

166 O. O. CECCARINI [J. s. M. p. E. 

signal, and, therefore, the beginning of each audio-frequency train 
will be somewhat chopped off. This effect is seldom apparent to the 
ear. The other effect is that the direct current will continue for some 
time after the audio signal has ceased. This produces at the end of 
each sound the familiar "swish" of surface noise, which is particu- 
larly noticeable with music of a staccato nature. The filtering action 
of the type of filter used with the noise-reduction circuit increases with 
the frequency, and any attempt to shorten the "time-constant" 
would result in degradation of the filtering action at low frequencies. 

If we were to impress the signal represented by equation 8 upon 
both elements of the push-pull valve with proper polarity, then upon 
reproducing such records through a push-pull circuit, properly ad- 
justed, nothing should be heard. Naturally enough, the conditions 
of critical balance are the same as mentioned before. 

If we now introduce a moderate amount of filtering, that is, reduce 
the filtering of the present standard noise-reduction circuit enough 
to permit us to achieve a time-constant of, say, Vsoo or VMO of a 
second, then the alignment of the slit would cease to be critical, 
since the high-frequency components of the noise-reduction are no 
longer present. 

We have now available all the elements of a good compromise, 
i. e.: the signal frequency is impressed upon the push-pull valve and 
is recorded in a class A fashion. The noise-reduction is obtained, as 
in the past, by rectifying the audio signal through a parallel circuit, 
and with filtering of moderate degree permitting the biasing current 
to reach its full value and to decay in a span of time as short as Vsoo 
or VHW of a second. This will eliminate the peculiar swishes so highly 
objectionable at present. 

Since the only components that need to be cancelled are those of 
low frequency contributed by the noise-reduction circuit and of ap- 
preciably reduced amplitude, the strict balance requirements of the 
push-pull reproducing circuit are materially reduced. 

The signal components being class A, will add numerically, and, 
therefore, the final signal is the numerical sum of the components 
from the two tracks. 

Since the push-pull circuit will cancel the even harmonics produced 
during the process of exposure by the asymmetry of the film character- 
istic and by the behavior of the light- valve, it is possible to obtain 
much cleaner records and greater amplitude range. 

It remains now to analyze the effect of imperfect balance upon 


the reproduced sound. The total input to each push-pull section of 
the light- valve under the new condition will be, respectively, 

E sin x + + 

( 2 , 

TT /() r(l - n 2 ) 

fi(x) = E sin 

where sin # is desired audio signal component, 2A/w is the direct- 
current bias, and 

represents the harmonic components from the noise-reduction rec- 
tifier not suppressed by the filter. The factor l//(w), which might 
be called the transmission characteristic of the filter, rapidly decreases 
with frequency ; in other words, the summation term above repre- 
sents substantially low-frequency components that will be cancelled 
out by algebraic summation of the push-pull output circuit. 

Now let us suppose that the tracks are not perfectly colinear, due 
either to a faulty set-up of the push-pull valve in recording or to faulty 
orientation of the slit in reproducing. Obviously this error can be 
only of very small magnitude and, therefore, noticeable only at high 
frequencies. If the audio signal is of high frequency then the audio 
signal component from one track will appear out of phase with respect 
to the other by a small angle 6. 

Therefore, we should have as the total output of the audio signal 
not 2E sin x but E sin x + E sin x\. Putting x x\ = 0, we have 

i n 

E sin x + E sin x\ = 2E sin ^(x + Xi) cos ^ 

But since 8 is constant for any one frequency we have 
E sin x + E sin xi = 2kE sin |(x + *i) 

(k = cos|)< 1 

In other words, the audio signal output will be of slightly lower am- 
plitude but without harmonic distortion. This, however, is the only 
effect that could be noticed, since the term representing the harmonic 


components from the noise-reduction rectifier is absent due to the 
action of the filter. 

Let us assume now that unbalance exists either in the reproducing 
circuit or in the illumination of the two tracks, or both. The ampli- 
tude of the audio signal component will be, say, R from one track 
and Ei from the other, and let EI < E or E! = K.E (K < 1). 
Then the total audio signal will be 

(E + Ei) sin x = E (1 + K) sin x < 2E sin x 
The apparent loss of volume will be 

Loss (db.) = 20 logic 1 ^_ R 

For the sake of clarity, if the unbalance is, say, 25 per cent, then the 
apparent loss of volume will be slightly greater than 1 db. This loss 
of volume will be present throughout the complete frequency range 
but without distortion to the wave form. 

Under this condition of unbalance the harmonic components repre- 
sented by the term 

will not be completely cancelled out, but will appear as the difference 
between the components of the two tracks. In the case of the numeri- 
cal example above it is evident that the uncancelled amplitude will 
be about 12 db. lower than the amplitude of each harmonic com- 
ponent alone, which in itself is low due to the attenuation character- 
istic of the filter. The results should, therefore, be still quite satis- 

Without entering into extensive discussions, on the basis of push- 
pull class A operation over a long period of time, it appears that this 
system offers the most acceptable compromise of several difficult 




Summary. The high-frequency camera of the Zeiss-Ikon Company has behind 
it twenty years of development. The original model, designed by H. Lehmann, 
appeared in 1917 as the Ernemann high-frequency camera. The principle is based 
upon optical compensation, to which end a reflecting drum with exterior mirrors was 
employed as compensating element. Films were exposed at a frequency up to 500 
pictures per second. 

The new model, which appeared upon the market in 1930, likewise depends upon 
mirror compensation, except that instead of the exterior mirrors a reflector drum is 
supplied with mirrors on the inside according to the patents of Professor Thorner. 

In this way an extraordinarily simple driving mechanism has been obtained, and 
a very compact form; with a capacity of approximately 60 meters of standard 35-mm. 
film, the size of the camera does not exceed the dimensions of a normal cine camera. 
The latest model permits an exposure frequency of about 1500 pictures per second. 

The camera is suited for use in technical photography of all kinds. It can be 
equipped with intermediate lenses for close-ups or with a supplementary distance 
tube for distance exposures. For photographing micro high-frequency films a par- 
ticular apparatus has been developed. 

To evaluate the exposures, a time-marking device is used, in which a glow-lamp, 
controlled by an electric tuning-fork, produces time records on the film at periods of 
Viooo sec. 

Whenever discussing the history of the motion picture we are 
accustomed, at least in Europe, to regard as the birthday of cine- 
matography the date on which the first photographically recorded 
series of pictures was shown before a large public audience; that is 
to say, the year 1895. 

Taking that date for granted, it seems strange that one special 
branch of cinematography, high-frequency or slow-motion cine- 
matography, is older than standard cinematography. In the annals 
of cinematographic history the high-frequency camera stands first 
and foremost, and the problems surrounding it constitute the original 
root from which the sturdy tree of cinematography has grown. 

* Presented at the Spring, 1937, Meeting at Hollywood, Calif. ; received 
July 9, 1937. 

** Technische Hochschule, Dresden, Germany. 




[J. S. M. P. E. 

In 1874 the French astronomer, Jules Janssen, created the photo- 
graphic revolver, by means of which he made a photographic record 
of the transit of Venus across the sun's disk, at a frequency of 50 to 
60 pictures a second. In 1877 the American photographer, Muy- 
bridge, made his celebrated photographs for studying the movements 
of the horse, which movements he was soon able to demonstrate by 
means of the so-called "wheel of life" (Zootrope}. 

These two demonstrations must undoubtedly be regarded as the 
beginning of high-frequency cinematography -that is, of the method 
of reproducing at retarded speed, pictures taken in rapid succession, 
an effect known in Germany as "time magnification." 

FIG. 1. Scheme of optical compensation by moving 
lenses (Jenkins 1894). The objective is moved parallel 
to film F, so that the picture of the object rests steady 
upon the moving film F. 

The French inventor, Jules Marey, in 1894 succeeded in making 
cinematographic pictures at speeds up to 120 a second by 
means of an intermittently moving negative photographic strip. 
His pictures of a falling cat, taken at that time at a frequency of 60 
pictures a second, are well known. Reproduction of the pictures 
by means of the Zootrope could be retarded six times. 

Apparatus with Continuous Movement. -The American inventor, 
Jenkins, deserves credit for having, in 1894, increased the taking 
frequency considerably by abandoning for the taking apparatus 
the principle of the intermittently moving picture strip and by 
adopting the continuously moving picture strip. By this method 
he obtained a filming frequency up to 250 pictures per second. The 
principle of this camera is illustrated in Fig. 1. By means of the 
film sprocket 5 the film F is drawn past the film-gate in continuous 

Feb., 1938] 



motion. To enable the lens to produce a well defined picture 
of the object, it must be moved in the same direction and with 
practically the same speed as the film. To achieve this, fifteen 
lenses were mounted on a rotat- 
ing disk. 

Another method of optical 
compensation by means of a 
reflecting drum was put on the 
market by the Frenchman, 
Reynaud. This apparatus was 
known as a Praxinoscope 
(Fig. 2). Instead of the simple 
reflecting drum, the French 
inventor, Mortier, in 1898 used 
an arrangement of square mir- 
rors to project uninterrupted 
moving pictures (Fig. 3). In 
this apparatus the compensat- 
ing mirrors lie between the film 
and the lens. The system of compensation may be described as 
interfocal compensation. The projected pictures show great lateral 
oscillation, as the virtual pictures lie in the axis of the mirror 

drum and oscillate. 

Thorner in 1900 overcame 
this trouble with his extrafocal 
mirror compensation (Fig. 4) 
whereby the separate mirrors 
of the reflecting drum lie out- 
side the focal length, between 
* the lens and the projection 
screen. The sprocket for film 
transport was rigidly connected 
to the reflecting drum m, and 
the perforated film traveled on 
the sprockets k on the circum- 

FIG. 2. Praxinoscope with mirror 
drum (Reynaud 1877) ; for inspection 
of moving pictures. 


FIG. 3. Revolving drum with 
square mirror (Mortier 1898); for 

ference of the drum. 

The illumination of the film 

pictures took place in the direc- 
tion of arrow i. The light rays, directed parallel by the objective 0, 
are thrown upon the prism p by the mirror of the reflecting drum m. 



[J. S. M. P. E. 

The diaphragms b, in the film drum, serve for separating the light- 
beams emanating from the various film pictures between the film 
and the lens. A variation of this form of mirror compensation was 
made in 1905 by the Austrian inventor, Musger, in which the rigid 
connection between the reflecting drum and the film transport 
sprocket was eliminated in favor of a gear-wheel coupling between 
these two parts. 

The External Reflecting Drum. Fig. 5 shows diagrammatically the 
range of optical compensation. The film F is fed continuously down- 
ward by the sprocket T. The objective lens 0, having a focal length 

FIG. 4. Extrafocal mirror compensation (Thorner 
1900, German pat. 124,932). The mirror drum m 
is outside the focal distance of the projection lens o. 

f, is opposite the film. In front of the objective is the reflecting drum 
with pivot A. In Fig. 5 only one of the mirrors, Si, is drawn in full 
detail, whereas the two laterally placed mirrors 2 and S$ are 
only roughly sketched. The object should be imaged in the direc- 
tion Z, so that the optical axis of the taking apparatus XYZ is 

The mirror drum is coupled to the transport sprocket T by geared 
wheels. The connection between the mirror drum and the transport 
sprocket must be such that if one point P on the film is moved over 
the distance s toward P', the mirror Si is revolved through the angle 
0-/25' so that the light-ray issuing from the object Z is concen- 
trated upon P'. The angle POP' then equals <r, or, in other words, 

Feb., 1938] 



the double of the revolving angle of the mirror drum, and it must 
have the value 

5 = / tan a 


Optical compensation requires that the rotational speed of the 
reflecting drum, as well as of the film transport, be constant. There- 
fore, the film distance 5 and the angle a are proportionate to time t; 
that is, 

s = cit a = ct (2) 

Instead of (1) we can substitute: 

cit = f tan (erf) (3) 

This conclusion must be valid for all values of /, or time, which is 
possible only when t is very small, so that the tangent may be sub- 
stituted by its argument. The optical compensation is correct, 


FIG. 5. Diagram of Thorner mirror compensation. 
During the revolution of the mirror drum SiSzSa, the pic- 
ture P of the object Z rests steadily upon the moving film 

therefore, only for small values of a. For large angles of rotation and 
large film distances (s] a skid is produced by the tangent discrepancy 
that leads to lack of sharpness in the direction of the film movement. 
The definition of the picture can be improved by narrowing down 
the film-gate by means of a sliding stop, which allows for each point 
of the film a very much shorter exposure time and a proportionally 
shorter exposure distance. In general, the exposure distance is made 
not larger than the height of the picture. In the design of the camera 
with the compensating mirror drum, it is important that the focal 

174 H. E. A. JOACHIM [j. s. M. P. E. 

length / of the lens, and also the number of mirrors n, be accurately 
determined from the picture spacing a. The angle a between the 
two mirrors Si and S z must be so calculated that the double distance 
PP' for the angle a (= a) is equal to the picture spacing a on the 

The beam of light emanating from the object Z, if reflected over 
two mirrors, must produce two images upon the film, the spacing 
between which is equal to the picture spacing a. Therefore, from 
equation 7, 

a = 2/ tan <x 
Now a = 2ir/n; so for a small angle a the result is 

To avoid the "skid" when the slit diaphragm in the film-gate is wide 
open (for larger exposure distances), various methods may be applied. 
We can either bend the film-guide toward the lens, or place a cy- 
lindrical lens in the path of light in front of the film-gate (as de- 
scribed in German patent 464,509). In this manner the tangent 
skid is avoided, and with optical compensation of this type well 
defined pictures will result. It goes without saying that the picture 
definition of such machines employing optical compensation can 
not be identical to that of instruments employing intermittent 
movement of the film, since the inaccuracy of mechanical gearing 
between the mirror drum and the film sprocket can never be en- 
tirely eliminated. 

The High- Frequency Camera of Lehmann. The first to succeed 
in constructing and satisfactorily operating a high-frequency camera 
with mirror drum was H. Lehmann, who began his first designs in 
1913. The first slow-motion camera of this type was put on the 
market in 1916 by the firm of Ernemann, under the name of Zeitlupe, 
meaning "time magnifier." 

Fig. 6 shows the interior of the camera. The reflecting drum is 
equipped with 40 mirrors. Behind the opening through which the 
light enters the camera is a mirror that throws the incoming light- 
beam upon the reflecting drum. The lens behind the latter concen- 
trates the beam upon the film strip located in the film-guide. Di- 
rectly below the film-guide is the sprocket for moving the film. 
Feed and take-up sprockets are located above and below the film- 
guide. The capacity of the film cassettes is 375 feet. The complete 

Feb., 1938] HlGH- FREQUENCY CAMERAS 175 

apparatus is shown in Fig. 7, fastened to a four-legged support. 
The construction of the table is such as to provide lateral and ver- 
tical movement. In the front panel are placed the handle for manual 
operation, the buttons for adjusting the sliding diaphragm and the 
iris diaphragm on the lens, and a ground-glass finder. On the left 
panel is the motor for the mechanical drive, which operates the 
apparatus by magnetic coupling. 

Before operating the Zeitlupe, the motor is started and at a given 
moment is connected with the machine by magnetic coupling. 
When operated by hand, a frequency of 300 exposures a second is 

FIG. 6. Lehmann's high-frequency camera (Ernemann Zeitlupe 1916); 
view of interior parts with mirror drum. 

attained; whereas when the apparatus is driven by motor a fre- 
quency of 500 can be reached. 

The slow-motion camera in this form was used during the War 
for various industrial, technical, and ballistic purposes. The more 
frequent use has been for photographing sporting events. The 
well known film by Dr. Frank, entitled The Wonders of Ski Sporting, 
was made with this camera. 

General introduction of the camera for cinematographic work 
was handicapped by its bulk and unhandy form, and by its great 
weight of some hundred pounds. In the meantime the cinemato- 
graphic industry had succeeded in making apparatus employing 
intermittently moving film strips, up to a frequency of 200 to 250 
pictures a second; so that the maximum speed of the old Zeitlupe, 



[J. S. M. P. E. 

500 pictures a second, did not offer such great advantages as to 
justify the higher price of the apparatus. 

Interior Mirror Drum. In order to improve the apparatus with 
regard to shape and efficiency, the design was entirely changed in 
1927 to 1929, and instead of employing Thorner's exterior mirror 
drum, the interior mirror drum invented in 1911 by the same in- 
ventor was used. The advantage was that the film could be placed 

in the interior of the mirror 
drum, reducing the dimensions 
of the apparatus to a very great 
extent. With a capacity of 175 
feet, the size of the taking 
camera was not very much 
greater than that of an ordinary 
motion picture camera. 

On the other hand, the interior 
mirror drum offers certain ad- 
vantages which depend upon its 
optical qualities, as Fig. 8 will 
demonstrate. Three parallel 
beams of light, 1, 2, and 3, 
coming from the object and 
falling upon three consecutive 
mirrors, 1, 2, and 3, of the mir- 
ror drum, are reflected by the 
interior mirror drum in a man- 
ner different from the way they 
are reflected by the exterior 
drum. In the first case, the 
reflected beams, /, 77, and ///, 
are divergent, as if they origi- 
nated from a virtual point inside the drum. With the interior mir- 
ror drum the rays, I, II, and ///, reflected by three consecutive 
mirrors, 1, 2, and 3, seem to be concentrated in such manner as 
to meet at the point Q, a sort of focal point outside the drum. 

With the taking lens system placed at Q, the beams /, //, and ///, 
originating from three different mirrors, will pass through the ob- 
jective. With the exterior drum, however, the only beam of light 
that passes through the objective is the beam II; the other beams, 
/ and IU, falling outside the taking lens. It therefore follows that 

FIG. 7. 

Lehmann's high-frequency 
camera (1916). 


the light-beam originating from the interior mirror drum is of greater 
intensity than that originating from the exterior mirror drum. 

New Type of Time Magnifier or Zeitlupe The construction of the 
new type of "time magnifier" is shown by Figs. 9 and 10. Fig. 9 (a) 
shows the apparatus viewed from the front when open, and Fig. 9(6) 
designates the various parts. In the interior part of the mirror drum 
S the film is contained in a double magazine C, with the take-up 
reel Ki and the feed reel Ki. The film leaves the magazine at 
the film slit R\ and returns into it through the slit JR 2 . Sprocket T 
moves the film in the direction of the arrow to the take-up reel K^ 
while N acts as hold-back sprocket. Between the film channel RI 
and the transport sprocket T, is located the film guide B, with the 

FIG. 8. Optical scheme for exterior and interior mirror 
drum ; the beam of light is reflected by consecutive mirrors 
1, 2, 3, in the direction I, II, III. Interior mirror drum 
with "focal point" Q. 

exposure gate into which the light-beam is concentrated by the taking 
lens O after having been reflected by the mirror drum. There are 
two additional mirrors, one located between the light entrance and 
the mirror drum, the other between the lens and the film-guide B. 
The lens is of the Ernostar type //1. 9. Due to loss of light on the 
surfaces of the intermediate mirrors, the aperture of //1. 9 is re- 
duced to practically //2.7. We have previously shown that the 
focal length of the taking lens depends on the number of mirrors 
on the drum. Therefore, the focusing of the camera can not be done 
by displacing the lens in the direction of the optical axis. It must be 
done by introducing secondary lenses, which are placed in the light 
entrance aperture of the apparatus. Without these secondary lenses 
the camera is focused at infinity. 


[J. S. M. P. E. 

Fig. 10 is a cross-section of the camera, looking from the rear left 
to the front. The driving motor M is on the outside of the housing 
G, with its shaft W supported by the bearings LI and L 2 . The 
mirror drum 5" is attached to the end of the motor shaft. The film 

FIG. 9. (a) Upper: High-frequency camera, 
new type 1930; (6) lower: view of inside 
parts with 175-ft. film magazine. 

sprockets T and N, as well as the take-up reel Kz, Fig. 9, are driven by 
gears on their shafts engaging directly with the inside gears of wheel E. 
By a magnetic coupling U, the wheel E can be coupled to the shaft 
of the motor. Two brushes, V, serve to feed the current to the 
magnetic coupling. In order to start the camera, the motor, with 
mirror drum and magnetic coupling device, is first of all brought 

Feb., 1938] 



to a high revolving speed. At the moment of the exposure the mag- 
netic coupling is actuated so that the film quickly reaches its con- 
stant speed. At that moment, the compensating mirrors come 
into full action, so that the pictures on the film are perfectly sharp. 

The construction of the new type of "time magnifier" is such that 
under all conditions a frequency of 1500 pictures a second can be 
reached. Fifteen to twenty feet of film are wasted before the coupling 
is effective. A frequency of 1500 pictures a second means a film 
speed of 70 miles an hour. 

Time-Recording Device. For research or testing purposes, it is 
necessary to know the exact speed of the film; that is to say, the 
frequency of the pictures. For this purpose a time-recording device 

Li V U 

FIG. 10. Cross-section of new type of cam- 
era, showing magnetic coupling of mirror drum 
with motor shaft. 

has been provided. It consists of a glow-lamp for producing re- 
cording-light marks upon the side of the film. The light is checked 
by an electric tuning fork having a frequency of 1000 cycles per 
second. Thus the edge of the film shows a time record in Vioooths 
of a second, so that the picture speed can be easily calculated. 

Microphotographic Time Magnifier. As soon as a "time magnifier" 
for ordinary cinematography had been developed, the apparatus 
was applied to microphotographic purposes, especially in zoological 
research work such as bacteriology. In all research work pertaining 



to microorganisms, most movements to be studied are of such high speed 
that ordinary cinematography does not suffice to show the motion. 

The difficulties encountered were to get adequate illumination and 
to provide convenient means of observation before and during 
the exposure. The apparatus used for the purpose is diagrammed 
in Fig. 11. 

The lamp is an 80-ampere arc lamp with high-intensity carbons. 
The light is concentrated by a condenser upon the mirror of the 


Detachable Se J^ ary 
Screen / Lens 




Time x 










^Arc Lernp^ 

Iris > 

Foot Switch 


FIG. 11. Diagram of microphotographic time magnifier. 

microscope. Between the condenser and the microscope are in- 
serted an iris shutter and a water-cooling jacket, to protect the micro- 
organisms in the microscope against destructive heat. 

Above the ocular of the microscope a rectangular prism serves to 
reflect the light-beam in the direction of the "time magnifier." The 
microscope has been especially designed by Carl Zeiss. The ocular 
lens has been removed, so that the magnifying lens consists only of 
the microobjective, which has been corrected to the longer picture 
distance. To provide for proper adjustment, the picture itself is 
imaged upon a detachable screen in front of the camera. For this 
short distance from the camera the secondary lens system must be 
used. To have the object under continuous observation, an auxiliary 
ocular has been inserted for the light-beam between microscope and 
reflecting prism. The camera has been used for various zoological 
purposes, and it has been possible to make exposures up to 1000 
pictures a second with an enlargement of 600 to 700 times. 



Summary. // has been shown that the median grain diameter of a developed 
photographic layer is a logarithmic function of the ratio of specular to diffuse density 
(Collier quotient}. Also, it is possible from the Collier quotient to evaluate the 
number of grains per unit area, the mass of silver per unit area, and obtain a value 
that parallels the graininess of the photographic deposit. 

As far back as 1890, Abney recognized 1 that a developed photo- 
graphic layer not only absorbs a portion of the incident light but 
also scatters some of it. The amount scattered determines the 
difference between the diffuse density D# and the specular density 
D" of the layer. The diffuse density is obtained by measuring 
the total emergent light and the specular density by measuring 
only those emergent rays that are approximately normal to the layer 
surface. Thus the diffuse density D# will always be equal to or 
smaller than the specular density D". The concept of specular and 
diffuse density may be explained by Fig. 1. 

Fig. \(A) indicates a developed photographic plate with a beam 
of light entering it normal to its surface. This is scattered by the 
developed emulsion. If the light-measuring device is close to the 
film, in the position shown, all the scattered light will be intercepted 
and measured. The density calculated from this light value is the 
diffuse density. In Fig. 1(J5) the same layer is illuminated as above, 
and likewise scatters part of the light. Here, however, the light- 
measuring device is placed so as to intercept only the unscattered 
light. From this measurement the specular density is computed. 

Callier 2 in his original paper mentions that the ratio of the specular 
to the diffuse density D"/D#, is closely related to the grain size and 
increases with it. This ratio, expressed as D"/Df = Q, has come to 
be known as the Callier quotient. Later Threadgold 3 investigated the 

* Translated and read by K. Famulener, Agfa-Ansco Corp., Binghamton, 
N. Y., at the Fall, 1937, Meeting at New York, N. Y. Received October 4, 1937. 
** I. G. Farbenindustrie Film Fabrik, Wolfen, Germany. 




[J. S. M. P. E. 

relation between the graininess and the scattered light, and, follow- 
ing Renwick's suggestion, 4 employed the Callier quotient. 

The following work was undertaken, first to give quantitative 
value to the qualitative relation between the grain size of photo- 

light- measuring 
* " device 

FIG. 1. Diagram illustrating concept of 
04) diffuse density and (B) specular density. 

graphic materials and the amount of light they scatter, and second, 
to show several possible applications of a measuring method based 
upon the Callier effect. 

In order to determine the diffuse and specular densities the ar- 
rangement shown in Fig. 2 was used. This apparatus has been 

A , A 2 

FIG. 2. Diagram of apparatus used for determining 
the Callier quotient. 

thoroughly described by H. Brandes. 5 The small square aperture at 
A i is illuminated by the light-source L and the condenser C 2 - The 
lens at C\ images this aperture on the opal glass O through the aper- 
ture A 2 . The photocell M of a densitometer is located behind 0. 
If the developed film is placed at At, only the light passing through 

Feb., 1938] THE CALLIER EFFECT 183 

approximately normal to the plane of the film will strike the photo- 
cell M. Thus the specular density is determined. 

If, on the other hand, the film is placed in contact with the opal 
glass at O, all the light passing through it will be measured by the 
photocell M, and the diffuse density determined. If the photocell 
is very close to O, the opal glass may be eliminated. 

The value of the specular density and, therefore, the Callier 
quotient Q, depends, to a great extent, upon the optical arrange- 
ment and the color of the light. For different values of the angle 
between the optical axis and the extreme rays passing through A% 
and entering O (that is, the angle a), a layer of medium grain size 
with a diffuse density of 0.50 has the specular density values shown 
in Table I. 


Values of D" when D# = 0.50 

( Degrees) 

3.0 0.69 

0.57 0.79 

0.29 0.87 

If a is kept constant at 0.57 degree and the wavelength of the light- 
source varied, the results shown in Table II are obtained : 

Color D" 

Infrared 0.85 

Red 1.09 

Green 1.15 

Blue 1.22 

It is therefore apparent that the smaller the value of a and the 
shorter the wavelength of the light-source, the greater is the specular 
density and, necessarily, the Callier quotient. 

In the following work, all measurements were made with a = 
0.57 degree; that is, the ratio of the opening at Ai to the distance 
from Az to O was 1 : 100. Also, all measurements were made with a 
normal low- voltage lamp and a caesium cell with high infrared sensi- 
tivity (the peak at 8600 A). Since the color of the light greatly 
influences the values obtained for the specular density, all determi- 
nations, in order that they may be comparable, must be carried out 
with the same colored light and on similarly colored silver deposits. 



[J. S. M. p. E. 

The results obtained in the following work refer only to those layers 
that contain uncolored, black, developed silver grains. 

It might be well at this point to make it clear that only the grains 
actually in the layer cause scattering of the light. Grain images 
printed from a negative on to a positive do not affect the light- 
scattering properties of that positive. If, for example, a negative 

0,05 0.10 0.15 0.20 0.25 O.3O O.35 


FIG. 3. Relation between average grain di- 

ameter (d) and Callier quotient 



having a Callier quotient of 2.0 is printed upon a grainless positive, 
the Callier quotient of that positive will still be 1.0, because there is 
no substantial graininess in the positive material. If, however, the 
print is examined, images of the negative grains will be seen. 


First the relation of the Callier quotient Q to the grain size, that 
is, the median grain diameter d, was investigated. For this work the 
median grain diameter d of different kinds of film was determined 
according to the method of Schaum and Bellach. 6 This is done by 
diluting the developed emulsion and spreading it out so as to have a 
layer not more than one grain thick. A photomicrograph is made of 
this and the average grain diameter determined by measuring the 
area of the projected grains and assuming them to be circular. The 
diameter is computed in microns. 

The Callier quotient of the initial densities (i. e., the densities 
before the emulsion was prepared for photomicrographic determina- 
tion) was also determined and it was found that the grain diameter 

Feb., 1938] THE CALLIER EFFECT 185 

d and the quotient Q were related to each other independently 
of the densities. (See Fig. 3.) 

However, it is possible that the Callier quotient varies not only 
with the grain size, but also with the emulsion thickness and the 
grain distribution. In order to settle this question, artificial silver 
grain emulsions of uniform grain size and varying thickness and 
density were prepared. This was done by melting a developed 
emulsion, thinning it with gelatin, and coating it in various thicknesses 
and concentrations on glass plates. 

If the value of the Callier quotient depends upon the layer thick- 
ness, or the density, then each of these coatings would give a different 
value for Q. As Table III shows, the Callier quotient Q is constant 
in all cases. In other words, for silver deposits with a uniform 
median grain diameter, Q is a constant independent of the layer 
thickness, density, and grain distribution. 


Callier Quotient of Artificially Prepared Silver Grain Emulsions of Similar Grain 


Coating Thickness Density Callier Quotient 

(Microns) D" Q - -^ 

15 0.23 .29 

15 0.29 .31 

15 0.36 .30 

15 0.46 .28 

15 0.60 .30 

15 0.75 .29 

15 1.14 1.30 

30 0.42 1.30 

30 0.57 1.30 

30 0.72 1.29 

45 0.65 1.31 

45 0.86 1.33 

45 1.14 1.30 

Table III justifies the relation shown in Fig. 3. Since the curve 
is a straight line passing through the origin, its general equation 
will be 

d=ClogQ (1) 

In this particular case, which applies to the experimental condi- 

186 J. EGGERT AND A. KUESTER [j. s. M. P. E. 

tions already outlined, the value of C, which is a characteristic of the 
apparatus, is 6.8. 

We have, therefore, the relation between the grain size and density 
in parallel and diffuse light : 

d = C log D' - C log Z># = C log ^ = C log Q (2) 

It must be emphasized that only the light reduction (or density 
increase) caused by the silver may be considered. If, for example, a 
film has a gray base, the density of the base must be determined 
separately and subtracted from the values for the specular and diffuse 
density. Moreover, if, besides the silver, a light-scattering medium 
(such as a line-screen or matte surface) is present, no determination 
is possible by the above method. In general, the absorption or 
scattering effect of a clear, colorless film or plate may be neglected. 


It is possible from the general equation given above to determine 
the number of grains Z per unit area of a developed photographic 
layer. The relation between the diffuse density, the number of 
grains, and the grain area is as follows : 7>8 

where A equals the median area of the projection of a single grain, 
and is considered circular. Then 

Since the median grain diameter d and the diffuse density D# may be 
determined from density measurements, Z is easily obtained as 
follows : 

Z^-t (5) 

(2.3) (4) (Dft _ 2.93 Df 

Table IV compares the value of Z obtained from the density mea- 
surements and by actual count. 

Feb., 1938] THE CALLIER EFFECT 187 


Determination of Number of Grains per Square Centimeter of Emulsion Surface 

Case D# Z Calculated from D# and D" Z Observed 

1 0.53 2.7 X 10 8 2.8 X 10 8 

2 0.47 1.7 X 10 8 1.7 X 10 8 

3 0.42 1.5 X 10 8 1.6 X 10 8 

4 0.58 1.2 X 10 8 1.1 X 10 8 



A third application of the Callier quotient is the determination of 
the amount of silver per unit area of a developed photographic layer. 

From numerous measurements on various kinds of emulsions that 
have been processed in various ways, it has been established empiri- 
cally, that the amount of silver per 100 square centimeters of surface 
depends upon the density and the grain size, while it is independent 
of the type of emulsion, the developer formula, and the time of de- 
velopment. 9 Instead of the amount of silver M per 100 cm. 2 , the 
photometric constant as denned by Hurter and Driffield may be used. 

Photometric Constant = P = (7) 

From this it appears that P is directly proportional to the median 
grain diameter. This relation is shown in Fig. 4. 

The amount of silver per 100 cm. 2 of surface may be determined 
in the following manner: Df and D" are measured and the median 
grain diameter d determined by equation 1. Knowing d, the photo- 
metric constant P is determined from Fig. 4, graphically. Re- 
arranging equation 7 we obtain 

M = P D# (8) 

so by multiplying the diffuse density by the photometric constant 
the amount of silver per 100 cm. 2 of emulsion surface is obtained. 
It is evident that such a determination does not give as accurate 
results as an analytical method. The inherent errors of density 
measurement cause variations that depend upon the density measured. 
The lower the density and the smaller the grain diameter, the greater 
is the error. With densities that are not too low (greater than 0.4), 
the average accuracy is =*=5 per cent. This has been discussed 
thoroughly elsewhere, 9 and it has been shown that the errors with 
this method correspond to the variations that are due to errors in 
determining the density. 

188 J. EGGERT AND A. KUESTER [j. s. M. P. E. 


It was of interest to learn whether any relation exists between 
the median grain diameter and the graininess. By "graininess" 
is meant that apparent graininess that one sees when examining a 
developed photographic layer under low magnification, or that 
shows visually in an enlargement. It is known that the median 
grain size decreases with an increase of density. 4 








FIG. 4. Relation of photo- 
metric constant (P) to median 
diameter d. 

FIG. 5. Relation between 
median grain diameter (d) and 
diffuse density (D#), for various 
times of development. 

On the other hand, it is also known that the graininess at first 
increases with the density, reaching a maximum at approximately 
D# = 0.30 to 0.40 and then decreases. 10 If the grain size and the 
graininess of several emulsions are to be compared, the comparison 
must be carried out at the same density. For the determination 
of the grain size of a developed photographic layer with the Callier 
quotient Q, we refer to the diffuse density D# = 0.50. The grain 
size is arbitrarily taken as 100 times the log of Q for a layer with 
a diffuse density of 0.5 and is indicated by K. 

K = 100 log ^ 

when D# = 0.50 
The grain size K was determined for a large number of different 

Feb., 1938] THE CALLIER EFFECT 189 

films from light-scattering measurements to see whether a parallel- 
ism existed between this and the graininess apparent when the film 
was sufficiently enlarged. In all cases layers with a diffuse density 
of 0.5 were used. On film developed to a diffuse density of D# = 
0.50, K was first determined. Then photomicrographs were made and 
enlarged 250 diameters, the density and contrast being kept equal. 
These photomicrographs were arranged in their order of visual 
graininess by several observers. The results are shown in Table V. 


Photomicrographs of D# = 0.50 Arranged according to Increasing Visual Graininess 

Observer Observer Observer Observer 

12 34 

18 18 18 18 

21 21 21 21 

26 26 26 26 

29 29 29 30 
32 32 28 28 
32 30 32 32 
28 28 30 32 

30 32 34 29 
34 34 32 34 
36 36 36 36 
38 38 38 38 

It is noticeable that the observers in general arrange the films in 
the same sequence as the increasing value of K. We may conclude 
that, with negatives of equal density, the grain size and graininess 
are approximately parallel. 

That the visual graininess of prints, made under the same proc- 
essing conditions, is parallel to the K value of the negative may be 
shown by the following tests : 

Pictures were taken of the same subject under the same con- 
ditions with ten different kinds of film. In each case a range of 
exposure was used. Negatives of approximately the same density 
were enlarged 10 diameters in a condenser enlarger, under identical 
conditions. The exposure and development of the enlargement were 
selected in each case so that the most favorable impression of the 
picture with regard to contrast and density was obtained. 

These enlargements were given to several observers for visual 
evaluation of the graininess. Table VI gives the order assigned by 
the different observers as well as that assigned by the same ob- 
server on three different days (indicated by a, b, and c). 



10-Diameter Enlargements Arranged in Order of Increasing Graininess 


/ Observer Observer Observer Observer 

a b c 2 3 4 5 

18 18 18 18 18 18 18 

21 21 21 21 21 21 21 

26 26 26 26 26 26 26 

30 29 28 28 28 29 29 

29 30 29 30 30 30 30 

28 32 30 29 29 28 32 

32 28 32 32 32 32 28 

32 32 32 32 32 32 32 

34 34 36 34 36 34 36 

36 36 34 36 34 36 34 

Here also a parallelism is found between the K value of the nega- 
tive and the visual graininess of the print. 


1 ABNEY: /. Soc. Chem. Ind., 9 (July 31, 1890), p. 722. 

2 CALLIER, A.: Zeitsch. Wiss. Phot., 7 (1909), p. 257. 

3 THREADGOLD, S. D.: Phot. J., 72 (1932), p. 348. 

4 RENWICK, F. F., AND BLOCK: Phot. J., 55 (Feb., 1916), p. 49. 
6 BRANDES, H. : Veroeff. Afga, 4 (1935), p. 58. 

6 SCHAUM, K., AND BELLACH, V. : Phys. Zeit., 4 (1902), p. 177. 

7 EGGERT, J., AND KUESTER, A.: Kinotechnik, 16 (1934), p. 127. 

8 ARENS, H., EGGERT, J., AND HEISENBERG, E.: Veroeff. Agfa, 2 (1931), p. 28. 

9 EGGERT, J., AND KUESTER, A.: Kinotechnik, 18 (1936), p. 381. 

10 Ross, F. E.: "The Physics of the Developed Photographic Image," D. 
Van Nostrand Co., New York, N. Y. (1924), p. 25. 


MR. SHEPPARD: What method was used to determine the mean grain size 
from the point of view of the distribution of grain sizes an arithmetical mean 
of 100, or how many counts? 

MR. FAMULENER: I do not know how many calculations were made, but the 
method was purely arithmetical. The area of the individual grains on a photo- 
micrographic breakdown was determined by Schaum and Bellachs' method; 
that is, the areas were measured, and the grains were assumed to be circular, 
and then the grain size was determined from those values. 

MR. SHEPPARD: But the distribution of sizes was not determined? 

MR. FAMULENER: No; merely the average. The distribution was not 
plotted against the diameter. 

* The values of graininess, K, in Tables V and VI were determined with visual 


Summary. A brief resume of the accomplishments of Robert William Paul in 
the motion picture field. At a meeting of the Society of Motion Picture Engineers on 
October 11, 1937, Mr. Paul was elected an Honorary Member of the Society. 

The Honorary Membership Committee has the honor to present the 
name of Robert William Paul for Honorary Membership in the 
Society. This proposal has the unanimous approval of the Historical 


By his ingenious design of many instruments necessary to the 
development of the motion picture, Paul distinguished himself and 
enriched the history of this industry. He was one of the first pro- 

* Presented at the Fall, 1937, Meeting at New York, N. Y. 



ducers and exhibitors of motion pictures. During his association 
with the industry for sixteen years (1894-1910) his work embraced 
all branches of activity in the development of the motion picture. 
In 1894, when Paul became interested in this field, the status of the 
art was somewhat as follows : 

Edison and Dickson completed their first camera using Tollable film 
in 1888-89 and began making short lengths of picture (47 feet long). 
The studio where the bulk of these films was made was completed in 
February, 1892. Many films were produced for examination with 
the peephole kinetoscope between 1892 and 1895, when Dickson left 
Edison. The camera was so large and heavy that the pictures were 
all of vaudeville subjects. No satisfactory projector had been de- 
veloped by Edison or his coworkers up to December, 1895, when 
Edison learned of the Armat vitascope and shortly afterward wit- 
nessed a demonstration. 

Jenkins and Armat gave the first public demonstration of their 
projector (phantoscope) in September, 1895, at the Cotton States 
Exposition in Atlanta, Georgia. Armat then made important changes 
in the machine and subsequently remolded it to make it of commercial 
value. The improved projector, known as the vitascope, was used 
for a public exhibition in New York City, April 23, 1896. 

Lumire began his researches on a camera and projector in 1894, 
and demonstrated them in March, 1895, at an industrial conference. 
The first exhibition for which admission was charged was given De- 
cember 28, 1895, in Paris. 

The work described below under Sections 1-7, inclusive, is con- 
sidered, therefore, to be contemporaneous with the work of Lumi&re 
and Armat. It is concluded, therefore, that Robert Paul pioneered 
in the evolution of the motion picture camera and projector, and 
influenced greatly the development of the motion picture industry in 
Great Britain. Mr. Paul's accomplishments may be summarized as 
follows : 

(1) Designed and built (with Acres) a camera with cam- driven intermittent 
movement (1895). 

(2) Designed and built printing and developing apparatus. 

(3) Designed and built an improved camera with a modified Geneva move- 
ment (1895). The camera was light and portable, as compared with the Edison 
and Dickson cameras. 

(4) Designed and built a projector with 7-tooth wheel intermittent movement 
(1895-96). (For description, see English Mechanic, Feb. 21 and Mar. 6, 1896.) 

(5) Designed and built an improved model projector having a revolving drum 


shutter cut away on two sides and equipped to show lantern-slides as well as 
motion pictures. Used arc or limelight (Brit. Pat. No. 4686, Mar. 2, 1896). 
Projector and camera preserved in Science Museum, London. 

(6) Designed and built a three-slot star-wheel intermittent projector with a 
30-degree shutter having a light-to darkness ratio of 11 to 1 (1899) (Brit. Pat. 
No. 487, 1899). 

(7) Gave the first entertainment with the projector known as theatrograph, 
at Finsbury Technical College, London, Feb. 20, 1896. (Same date as the first 
showing of Lumiere's projector in England.) 

(8) Built the first motion picture studio in England used for commercial 
production at Muswell Hill, N. London (1899). (Described in Strand Magazine.) 

(9) Gave many exhibitions in and around London, and supplied pictures 
made under his direction for numerous exhibitions. 

(10) Made pioneer investigations hi trick photography and slow-motion 
studies of a scientific nature for distribution. 

(11) Made some of the earliest news or topical pictures, during 1896, in Por- 
tugal, Spain, and Egypt as well as in England. Pictures shown with Paul's 
projector on the S. S. Norman on a trip to South Africa, April, 1896, were prob- 
ably the first pictures exhibited at sea. Photographed the Prince's Derby hi 
June, 1896, and Queen Victoria's Diamond Jubilee hi 1897. Sent two cameras 
to the Boer War in 1899. 

Mr. Paul began his career as a manufacturer of electrical and 
scientific instruments in London in 1891. He gave up the motion 
picture work in 1910 to devote himself exclusively to his original 
business. In the intervening years since that date he has made a 
notable reputation for his skillful design of instruments, until his 
health forced his retirement a few years ago. 

J. I. CRABTREE, Chairman 


British Patents granted to R. W. Paul between the years 1895-1905, in- 

17,677/95 Kinetoscope Apparatus 

19,984/95 Exhibition on Entertainment 

4,168/96 Reproducing Stage Performances 

4,686/96 Projecting Kinetoscope Pictures on the Screen 

10,310/97 Exhibiting Animated Photographs 

486/99 Taking and Projecting Pictures 

11,997/99 Animated Photography 

14,372/00 Projecting Photographs 

26,747/01 Taking and Projecting Animated Photographs 

Election of Robert William Paul. At a meeting of the Board of 
Governors of the Society of Motion Picture Engineers on October 10, 


1937, in the Hotel Pennsylvania, New York, N. Y., Robert William 
Paul was proposed and unanimously approved for Honorary Member- 

In accordance with the provisions for electing honorary members 
the nomination was then placed before the General Society at the 
New York meeting on October 11, 1937, at which tune the nomination 
of the Board received the unanimous endorsement of the general 
Society. At that meeting, an account of Mr. Paul's work in cine- 
matography was read by Mr. G. E. Matthews, member of the Histori- 
cal Committee. 

On October 12th, Mr. Paul was notified by letter, from President 
S. K. Wolf, of his election to Honorary Membership. 




Summary. American motion pictures are maintaining immense popularity 
throughout the world, yet barriers and obstructions that tend to limit their sale con- 
tinue to be imposed abroad. Safeguarding and developing our film market abroad, 
"contingents'" taxes, and complex restrictions, which continue to be imposed in all too 
many instances, are some of the problems American producers must face. Some are 
legitimate from the standpoint of local interests, but others appear to be unreasonable. 
In certain cases our motion picture industry may be justified in taking a strong and 
positive stand with the object of bringing about the rectification cf unfair measures. 

The efforts of the Bureau of Foreign and Domestic Commerce to safeguard and 
augment American motion picture markets by supplying factual data and utilizing 
trade-promotive methods is covered. The Bureau's motion picture unit has recently 
been raised to full divisional rank, in recognition of the industry's importance. Such 
helps as it provides are especially vital at the present moment because our motion pic- 
ture producers and distributors are likely to find themselves puzzled, entangled, or 
thwarted by the ever-growing intricacy of the conditions that they face abroad; their 
continued success in foreign markets depends upon the functioning of a reliable in- 
telligence service. 

As motion picture engineers, the members of this^Society all have 
a vital and immediate interest in the vigorous maintenance of our 
foreign markets for American motion pictures. Those foreign mar- 
kets constitute a highly important field for the efforts of the industry; 
they play a significant role in its notable success. They can not be 
considered lightly; they must, on the contrary, be constantly cher- 
ished and cultivated and energetically safeguarded. Every endeavor 
along that line means much to a motion picture engineer because the 
opportunity for his advancement, or even the security of his position 
may be dependent in no small measure upon the industry's ability to 
hold these foreign markets which have furnished, up till now, such 
important sources of revenue. 

It is an easily demonstrable fact that any type of foreign legis- 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received Sep- 
tember 20, 1937. 

** Chief, Motion Picture Division, U. S. Bureau of Foreign and Domestic Com- 
merce, Washington, D. C. 


196 N. D. GOLDEN [J. s. M. p. E. 

lation that operates to retard or prevent the sale of American motion 
pictures abroad inevitably reduces the income of American distribu- 
tors which, in turn, naturally leads to curtailment of production, 
with a resultant laying-off of trained personnel not only at the studios 
but in the research laboratories as well. Thus, a strong wave of 
nationalistic sentiment in Central Europe, finding expression in re- 
strictive laws affecting motion pictures, or a movement toward con- 
trol and rigid censorship somewhere in Asia, or some inimical reaction 
in a Latin-American country, may contribute to an ultimate effect 
whereby a motion picture engineer would find less in his pay-envelope 
or might even be confronted by more severe emergencies. On the 
contrary, smooth and favorable expansion of the foreign markets 
might involve substantial heightening of his personal income and 

It is obvious, therefore, that the industry needs to maintain an 
attitude of the keenest and most alert watchfulness on all foreign 
developments. It needs all the detailed facts with regard to what 
is happening in the motion picture field abroad. It needs those facts 
promptly, and it is only fair to say that, in all such efforts as these, 
it needs the cooperation of the Commerce Service of the National 

Such needs were never more imperative than they are at the pres- 
ent moment, when world conditions are transitional, confused, and 
often muddled, when new situations develop with truly amazing 
suddenness, and 'the basic governing factors are frequently difficult 
to ascertain. Under circumstances such as these, our motion picture 
producers,and distributors are likely to find themselves puzzled, en- 
tangled, or thwarted by the ever-growing intricacy of the conditions 
that they face abroad. Their continued success in foreign markets 
depends upon the functioning of a reliable " intelligence service" to 
keep them fully informed on all the foreign facts and figures, the 
quotas, the limitations, the control boards, the taxes, the fostering 
of local competition, and the many other vital factors that bear upon 
their business. 

Such a service is of course supplied in considerable measure by the 
personal representatives of our larger motion picture interests, but 
the Government at Washington, through the Bureau of Foreign and 
Domestic Commerce, can assuredly be helpful. We believe that we 
are in a position to make an appreciable contribution to the safe- 
guarding and developing of motion picture markets throughout the 

Feb., 1938] FILM MARKETS ABROAD 197 

world. On July 1st of this year the Bureau recognized the great im- 
portance of the motion picture trade by raising this phase of our pro- 
motive work to full divisional rank. We now have a Motion Picture 
Division, eager at all times to serve the industry, and possessing the 
resources and equipment to do so with effectiveness. Quoting the 
words that our Director, Dr. Alexander V. Dye, recently applied to 
the entire Bureau, the Motion Picture Division aims to help the in- 
dustry "to avoid errors, avert losses, and solve problems to open 
pathways to wise planning, sound judgments, and ever-greater 
profits." That is the objective toward which we steadily strive. 

To aid in its attainment we have Foreign Commerce Officers in 
thirty-four foreign capitals and great commercial centers who keep 
the home office at Washington promptly informed of every change, 
development, tendency, or measure that may have any repercussions 
on the American motion picture business. The reports of these rep- 
resentatives, as well as those of the State Department's consuls, 
are reviewed, analyzed, and coordinated in the Motion Picture Di- 
vision and disseminated by it to the interested companies or persons. 
This is accomplished through the issuance of bulletins and circulars, 
through letters and personal interviews, and in other ways that seem 
best adapted to particular situations. Naturally there is often an 
opportunity for service by our district offices, such as the admirably 
equipped offices in Los Angeles and New York City. In key cities 
throughout the country our Bureau maintains twenty-four of these 
"service stations." 

As an example of the kind of data furnished by our field service for 
the benefit of the motion picture industry, there is now available a 
concise yet complete report from one of our Assistant Trade Com- 
missioners in London, presenting the details of the proposals for legis- 
lation on motion picture films recently submitted to the British Par- 
liament by the President of the Board of Trade, which is the govern- 
mental body corresponding to our own Department of Commerce. 
This report tells all about the proposed continuance of the British 
quota system, the differing provisions for distributors and exhibitors, 
the "cost test," the excepted classes, the ban on blind booking, the 
penalties for violation, and the numerous other recommendations 
that the British are likely to embody in their new legislation on this 
subject. In this connection it appears that Britain will continue to 
be an eminently worth-while market for the splendid films created 
by our engineers and artists. 

198 N. D. GOLDEN [J. S. M. P. E. 

The Motion Picture Division advises the industry on all the phases 
of film conditions abroad, including markets for educational and in- 
dustrial films and for motion picture equipment. Special statistical 
data, covering our exports and imports and the foreign production 
of films, are given to the trade each month, as are the reports of for- 
eign censorship boards. Bi-weekly we publish a Foreign Market 
Bulletin, covering a wide variety of motion picture news from the 
oversea markets. A press service of pertinent items goes each week 
to export managers, trade associations, trade papers, and newspapers. 

A significant function of the Division is its service in furnishing 
each month a bulletin on the latest production and distribution of 
non-theatrical films to home users, colleges, and schools. Equip- 
ment manufacturers, exporters, and associations are currently ad- 
vised on the foreign-market potentialities for all types of motion 
picture and sound-reproducing equipment. We are constantly pub- 
lishing information on competition abroad, the number of films dis- 
tributed by countries over given periods, the number of theaters in 
the different foreign territories, and similar matters that ultimately 
mean much, in dollars and cents as well as in prestige, to the men 
and women who are making pictures here in the United States. 

Let us look briefly at the Division's major publication, the Review 
of Foreign Film Markets, of which the number covering the year 
1936 appeared some months ago. In this large processed bulletin, 
comprising about 180 pages, the situation in nearly every country 
of the globe is shown succinctly but in some detail. Here, to men- 
tion only a few facts out of many, it is found that film censorship in 
Nazi Germany has been greatly intensified and is proving most 
vexatious to motion picture importers; that more interesting sce- 
narios, with better international value, are now being used by French 
producers; that filmgoers in the Dominican Republic are voicing 
vigorous objections to dubbing the native language into the pictured 
mouths of our American stars ; that 73 per cent (by footage) of the 
pictures imported by India are of American origin; that in Mexico 
only seven companies regularly produce motion pictures ; that there 
are 38,190 seats in the picture theaters of Montevideo; that import- 
ers of films to Latvia pay a special tax of 0.15 lats per meter for the 
benefit of the Latvian Culture Fund; and so on through the list of 
countries and the long categories of devices, regulations, and local 
measures that affect the market for our films abroad. 

American pictures continue to maintain their immense popularity 

Feb., 1938] FILM MARKETS ABROAD 199 

abroad. Theatergoers are enthusiastic about them in Norwegian 
coastal cities and in the high Andean capitals, in the teeming Oriental 
centers and in the sophisticated, metropolitan theaters of western 
Europe. That is only natural, of course, when one considers the en- 
gineering genius, the artistic ability, and the business acumen that 
go into their making. Foreign peoples like them and demand them, 
because they realize their incomparable excellence. And yet the 
barriers and obstructions and hobbles that tend to limit their sale 
continue to be imposed abroad. Contingents and taxes and complex 
restrictions continue to be slapped on, in all too many instances. 
Some of these are legitimate enough, from the standpoint of local in- 
terests; others, however, appear to be inherently unreasonable. In 
certain cases our motion picture industry may be justified in taking 
a strong and positive stand with the object of bringing about rectifica- 
tion of unfair measures. We need not be unduly hesitant. Our 
producers and distributors can afford to make effective their opposi- 
tion to merely narrow-minded or punitive practices, while at the same 
time conforming readily to rational and moderate requirements 
abroad. In any such stand they will have the backing of one momen- 
tous factor namely, the avidity of foreign audiences to see and hear 
our magnificently entertaining films. 

In any event, it is the unremitting effort and steady purpose of 
the United States Department of Commerce to safeguard and develop 
our motion picture markets abroad to the greatest possible extent 
through the instrumentality of factual data and a variety of tried 
and proved trade-promo tive methods. 


MR. CRABTREE: Can you outline briefly the pros and cons from the American 
standpoint of establishing studios in England by the motion picture producers? 

MR. GOLDEN : In addition to what you have already read in the trade-press 
we have received reports from our trade commissioners in London indicating 
that American companies are either leasing or buying outright certain studios 
and space in studios. This is a precautionary measure, caused by the issuance of 
the White paper, which deals with the revision of the quota system that expires 
in 1938. It contains proposals that will have to be passed upon by the Parliament. 

MR. CRABTREE: A certain percentage of the pictures must be of British pro- 
duction, is that not right? 

MR. GOLDEN: Twenty per cent, at the present time. 

MR. CRABTREE : If the pictures are made by an American producer in England, 
are they classed under the quota? 

MR. GOLDEN: Provided they meet the terms of the quota act, which requires 

200 N. D. GOLDEN 

that a certain percentage of the employees or the personnel be British subjects; 
75 per cent, I believe, must be British subjects. 

MR. RICHARDSON : You have said that many objections were raised to showing 
American films. Do any of the objections relate to the action in the picture? 

MR. GOLDEN: That depends upon the attitude of those in charge of permitting 
films to come into the countries. Some countries, based upon their censorship 
laws, may .object to the slightest thing in the picture, to bar the picture from the 
market. We have such trouble in Czechoslovakia and in Germany. If the man 
who turns the crank of a camera or the producer happens to be of the Jewish 
faith, that is sufficient ground to bar the picture from the German market. The 
American motion picture industry has been very careful in selecting its pictures 
for foreign markets. At the studios there are experts who are well versed with the 
censorship laws of the various countries, and situations that may not get by in 
foreign countries are quickly eliminated from the pictures. 

MR. RICHARDSON: What is the character of such objectionable items? 

MR. GOLDEN: Almost anything. It might be a question of morals, or it might 
be political. Perhaps the dictator of a country may feel that his subjects should 
not see a certain kind of picture. It may be one of a thousand different things 



Summary' A brief historical resume of the subject of bird sound recording is 
presented, including a discussion of the idea behind the work and its usefulness to 
students of ornithology. Graphical methods of recording songs are mentioned, as 
well as some early attempts at phonographic recording. 

Problems of recording bird songs in the field are discussed. The high frequencies 
of bird song; the necessity of working at relatively great distances from the subject; 
wind and other noises; the need for portability and simplicity of equipment, all com- 
bine to increase the difficulties of the work. The use of parabolic concentrators in the 
work is discussed and it is concluded that the frequency distortion introduced by this 
type of pick-up device is more tolerable for bird song recording than it would be for 
records of human voice or music. 

As in the beginnings of many things, we shall probably never know 
who first attempted to include the song of a bird in the repertoire of a 
phonograph, and we may not even know the names of many of those 
who were associated with the work. The problem, however, seems so 
obvious to the ornithologist that it must have been thought of many 
times before the literature recorded the event. 

To the naturalist, song is no less a part of a bird's make-up than its 
form or coloring, although it has defied man in his attempt to record 
it for a much longer time. 

There are many records of attempts to record bird songs with musi- 
cal notations, 1 but since bird music rarely complies with man-made 
scales or time intervals, records of this type are so complicated that 
they are of little use to one who would like^to know what the bird 
sounded like. Saunders, 2 in this country, has devised a very work- 
able system of short-hand for writing down bird songs, and with a 
little study one may gain a fair idea of the song to be expected from 
any bird. Fig. 1 illustrates the use of this method in recording several 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received Sep- 
tember 30, 1937. 

** Laboratory of Ornithology, Cornell University, Ithaca, N. Y. 




[J. S. M. p. E. 

songs of the rare Kirtland's warbler, which is found breeding only in 
a restricted area in Michigan. 

It was twenty-one years after the invention of the phonograph that 
Sylvester Judd, 3 on November 15, 1898, demonstrated a phonographic 
recording of the brown thrasher's voice before the 16th Congress of 
the American Ornithologists' Union in Washington, D. C. While this 
was recorded as a unique feature of the meetings, and as a method of 
study of great promise, nothing was said to indicate the quality of the 
recording or the opinion of the audience. Mr. Judd's work was cut 
short by his untimely death, and I have found no further record of 



chert chert chert cher cjitcjitoui-wi 

ehu cW chu chS chS che wt-iowK 



(Prepared by H. U. Axteil) 

FIG. 1. Example of graphic method of recording bird 
songs devised by A. A. Saunders. 

attempts to record bird song phonographically until 1909, which Hein- 
roth 4 sets as the time he heard reproductions of North American bird 
songs at a public lecture in London. I have been unable to find the 
source of these recordings. The following year Heinroth states that 
he played several records of bird songs before the 5th International 
Ornithological Congress in Berlin, and specifically mentions a good 
reproduction of a nightingale's song. So far as I know, all these early 
recordings were of caged birds. 

With the advent of electrical recording about 1926, many sugges- 
tions were published 5 indicating the possibility of using the new tech- 
nic with birds, and in 1927 the RCA Victor Company published songs 
of several caged birds recorded in the Karl Reich Aviary in Bremen, 



Germany, and the song of a semi-domesticated nightingale from an 
English garden. 6 

In May, 1929, the Fox-Case Company sent out two men with equip- 
ment to Auburn, N. Y., to record sound pictures of some wild birds for 
a springtime release. After spending considerable time about Au- 
burn, with no success, they came to Cornell to enlist the help of Profes- 
sor A. A. Allen, to make the birds hold still. From his long experience 
with bird photography Dr. Allen was able to help them out, and that 
morning songs of three common birds were recorded synchronously 
with motion pictures. 

This experience interested us anew in the possibilities of recording 
bird voices, but we were unable to get started on the project until 1931. 

(Photo by A. A. Allen) 
FIG. 2. Camp in swamp for studying ivory-billed wood- 

In 1930 A. R. Brand enrolled in Cornell as a special student in orni- 
thology, and presently became interested in recording bird songs as a 
method of study. Since that time Mr. Brand has been actively en- 
gaged in this work and has financed the experimental work in this 
field in the Laboratory of Ornithology at Cornell. 

The problems encountered in bird sound recording fall naturally into 
three groups: (1) those imposed by field conditions such as need for 
portability and ruggedness; (2) problems connected with sound 
pick-up; (3) problems arising from the character of the sound we 
are working with. 

That it is very desirable that the equipment be portable and rugged 
is illustrated by our experience in recording the voice of the ivory-billed 
woodpecker in the swamps of Louisiana. Here we found this rare 

204 P. KELLOGG [J. s. M. P. E. 

bird nesting in the middle of a deep swamp at least six miles from any 
road on which an automobile could travel. Fig. 2 gives some idea of 
the difficulties encountered. Our equipment was not portable, ex- 
cept by truck, and the task of moving it to a new location by mule- 
power required three days. Under normal circumstances only the 
microphone is removed from the small truck, which serves as a field 
laboratory; but, when working with rare birds, trails often lead to re- 
mote areas where a recorder in a back pack would be much more use- 
ful than one in a truck. 

(Photo by A. A. Allen) 

FIG. 3. Water ouzel or dipper. Only by 
getting the bird close to the microphone could 
the roar of the rushing water be overcome and 
the song of the bird recorded. 

During the present summer (1937) the trail of the Leache's petrel, 
an oceanic bird, led to the islands of the Bay of Fundy. Being unable 
to get the truck conveniently to Kent's Island, where the birds were 
calling best, use was made of a small portable field amplifier and the 
short-wave radio station of the Bowdoin (College) Kent's Island Ex- 
pedition to transfer the sounds of the petrels across the six miles of the 
Bay of Fundy to the Island of Grand Manan, where the recording took 

The problems connected with picking up bird songs are most per- 
plexing. Ideally, studio conditions should* prevail; but birds are 
seldom at home indoors and critics are quick to detect the differences 
between the song of a free bird and that of a caged one. Even out of 


doors, studio conditions may often be simulated when one has the co- 
operation of the artist, but with birds this is a little too much to expect. 
Theoretically, many birds have regular song perches to which they go 
to sing, and it should be possible, by placing the microphone near such 
a perch, to record good sound. This is a good theory, but the times 
at which most birds return to their song perches are so unpredictable 
that in practice the method is seldom used. Fig. 3 illustrates the 
set-up for recording the song of the dipper or waterouzel of the West. 
Here the noise of the rushing stream frequented by this bird made it 

(Photo by A. A. Allen) 

FIG. 4. Thirty-two-inch parabolic concentrator being 
focused on ivory-bills by James Tanner. 

necessary to study the song habits carefully until we had practically 
learned every spot from which the bird could be expected to sing. 
Only in this way could the bird's voice be recorded above the sound of 
the water. 

In 1932 Peter Keane and the writer decided to build a sound con- 
centrator in an effort to achieve better pick-up. We were thoroughly 
warned by our engineering friends that the frequency characteristics 
of any reasonably proportioned reflector would make it useless for our 
purpose. 7 ' 8 However, we insisted upon going ahead with the idea, 
and constructed a parabolic concentrator 32 inches in diameter (Fig. 
4). Measurements of the concentrator under field conditions are 
approximately shown in the curve of Fig. 5. Theoretically, such a 



[J. S. M. P. E. 

response would be quite useless for any kind of recording, but for bird 
work two conditions make it at least tolerable, and this brings us to a 
consideration of the sounds with which we are working. Brand 9 has 
pointed out that the average frequency of bird songs he has studied 
is a little above 4000 cps., and in this region the gain from the use of 
the concentrator is about 22 db., which gives us a "magnification," 
thinking in terms of a field glass, of approximately 12.5 times. This is 
a distinct advantage. The second consideration that makes such a 
response tolerable is the relatively narrow frequency range of most 
birds. This has the effect of confining the distortion within relatively 
narrow limits, and even this may be somewhat further reduced by the 




FIG. 5. Response of 32-inch parabolic concentrator 
compared to response of microphone alone, under field 
conditions at a distance of 60 feet. 

tendency of most reproducing equipment to fall off with increasing 
frequency. For voice recordings or for ordinary music the distortion 
introduced by such a small reflector is very noticeable; but for bird 
sounds, listening tests show that recordings made with the parabola 
are preferable to those so far made with the microphone alone. How- 
ever, we feel that much of this preference may be due to the lower 
background noise with the concentrator rather than to the emphasis 
upon the higher frequencies. 

Two possible solutions to the attempts to make better recordings of 
bird songs in the open would be, first, to make more concentrated 
studies of individual species so as to place the microphone close to 
the bird; and, second, to use a larger parabola so as to concentrate 
more sound and give a better overall response. It is probable that 
the first suggestion would result in better quality when it could be 


achieved, but the parabola is so much more efficient that its use will 
probably continue, and sometimes it affords the only means of re- 
cording a song. 

In our recording work we are particularly anxious to record the 
songs of rare and vanishing species of birds before it is too late. In 
modern times the passenger pigeon, the heath hen, the great auk, and 
the Laborador duck have become extinct. No one has accurately re- 
corded the voices of these birds and they are gone forever. The 
ivory-billed woodpecker is on the verge of extinction. 

( The presentation -was concluded with the projection of sound pictures of birds and 
their songs, these records being part of the results of an expedition sponsored in 1935 
by Mr. Brand, Cornell University, and the American Museum of Natural History, to 
study and make records of vanishing species.) 


1 MATTHEWS, F. S.: "Field Book of Wild Animals and Their Music," G. P. 
Putnam Sons (New York, 1904), p. 262. 

2 SAUNDERS, A. A.: "A Guide to Bird Song," Appleton- Century (New York, 
1935), p. 278. 

3 JUDD, S. D.: "Gramophone Demonstration of a Brown Thrasher's Song," 
Auk, XVI (Jan., 1899), No. 1, p. 52. 

4 HEiNROTH, O.: "Gefiederte Meistersanger," H. Bermuhler (Berlin, 1936), 
p. 96 (plus three 10-inch disks). 

6 STABLER, H.: "Das Freiland- phonographieren von Vogelstimmen," 
Bericht des Vereins Schlesischer Ornithologen (Breslau), XII, p. 95. 

6 Victor Records, Nos. 20968, V-l, V-15, V-50, V-71, V-85, 22344. 

7 DREHER, C.: "Microphone Concentrators in Picture Production," /. Soc. 
Mot. Pict. Eng., XVI (Jan., 1931), No. 1, p. 23. 

8 OLSON, H. F., AND WOLF, I.: "Sound Concentrators for Microphones," 
J. Acoust. Soc. Amer., I (April, 1930), No. 3, p. 410. 

9 BRAND, A. R.: "Method for the Intensive Study of Bird Songs," Auk, n.s. 
LII (Jan., 1935), No. 1, p. 40. 


MR. BRADLEY: Did you make any attempt to record sounds of birds by the 
phonographic method only, without the picture? 

MR. KELLOGG: Yes; many of the sounds have been transcribed upon disks 
and published as illustrative material for a book, by Mr. Brand, who financed the 
work that I mentioned. With a good phonograph we can get fairly good results. 
The frequencies average about 5000 cps., and with the 78-rpm. speed and fine-line 
recording we do not obtain good results on an ordinary acoustical phonograph. 

MR. BRADLEY: Recently, the American Foundation for the Blind offered The 
National Archives some phonograph records, made by Columbia University, of 
wild birds and their songs. The National Archives Act states that we may 
accept motion pictures and sound recordings pertaining to historical activities 


of the United States. The question has been raised as to whether or not recording 
birds' songs constitutes an historical activity and hence whether or not such 
recordings would be admissible under the provisions of the Act. I should be 
interested to have your opinion. 

MR. KELLOGG: Within the memory of man in the United States about six 
birds have become extinct, and no one will ever know what they sounded like, 
except from graphic descriptions of them. Personally, I think it should be 
regarded as part of our national history to know what kinds of birds, what kinds 
of animals inhabited this country before man came here or during his early stages 
here. It seems to me that would be a rather legitimate inclusion for The National 

The records of which you speak were made in our laboratories at Cornell 
that is, the editing was done there and Mr. Brand wrote the continuity that 
goes with them. He is now producing a second book. But I suppose if we get 
out victrola records that are good enough and if the master will stand up, whether 
The National Archives keeps them or someone else, we shall have permanent 
records of them. 

MR. POPOVICI: Has any attempt been made to improve the response of the 
parabolic microphone with networks? 

MR. KELLOGG: Yes. But as soon as we try to suppress the high end of the 
curve, where the response is so great, we very soon find ourselves back almost at 
the point from where we started. The most interesting attempt has been with 
resonators, to bring up the response at the low end; but I think the solution is a 
larger parabola having a less steep response curve. 

MR. CRABTREE: How do you focus the reflector, by trial and error? Also 
can you tell us something about your telephoto equipment? 

MR. KELLOGG: The pictures were made with a 17-inch lens, the telephoto 
equipment on the motion picture camera. Focusing is difficult. Sound-waves are 
not as sharply focused as light-waves, and the simplest way we have found to 
focus the parabola is to aim it at the sun. Out in front, at the focal point, the 
parallel rays of the sun concentrate into a very hot spot. As the subject is closer 
than the sun, the focal point moves out a little bit, but that has merely the effect 
of broadening the spot, and in almost every instance it will cover the diaphragm 
very well. Experience shows that it is desirable to throw the concentrator 
slightly out of focus so as to have less high-frequency response. 

We have a little telescope on the concentrator, placed right along the side. 
(The man in Fig. 4 is shown sighting through it.) At 5000 cps., which is about 
the average of bird songs, we get a gain of about 22 or 23 db., which is a very 
appreciable gain, I assure you. It is a very powerful instrument. 



Summary. Until such time as all theaters are equipped with new and modern 
improved equipment, methods must be pursued that will allow wide-power-range 
films to be reproduced to best advantage in theaters having equipment capable of such 
reproduction, and not, at the same time, penalize theaters having equipment incapable 
of handling the wider range. During the past year several major producing com- 
panies have made available to the theaters prints of two types: (1} The "Regular" 
release print with ordinary volume range; (2) " Hi -Range" and" Lo-Range" prints. 

The "Hi-Range" prints have a range of 50 db. The volume of the "Lo-Range" 
prints may correspond to that of the "Regular" prints, or may be recorded to play 3 or 
4 db. above the particular studio's average. In other words, any production issued on 
"Regular" prints will be distributed completely on one type of print, while any pro- 
duction available on "Hi-Range" prints will be available also on "Lo-Range" prints as 
well. As more theaters become converted to equipment capable of the higher ranges 
the practice of issuing "Hi-Range" and "Lo-Range" prints will be rapidly extended. 

Instructions are given for using the various kinds of prints in theaters, and curves 
show the recommended amplifier output in terms of theater area, cubical contents, 
and number of seats. 

Since the addition of recorded sound to motion pictures there has 
been continual improvement in the quality of sound recording and 
sound reproduction which has been particularly marked within the 
past year. 

The Research Council of the Academy of Motion Picture Arts & 
Sciences hopes by means of the work of its Committee on Standardi- 
zation of Theater Sound Projection Equipment Characteristics to 
give to the exhibitor throughout the country a more intimate picture 
of the aims of the producers in attempting to obtain a more natural 
sound recording. As the quality of recorded sound is improved and 
its naturalness is increased, it is necessary, in order to obtain the 

* Received October 7, 1937; presented at the Fall, 1937, Meeting at New 
York, N. Y. 

'* Chairman, Committee on Standardization of Theater Sound Projection 
Equipment Characteristics of the Research Council of the Academy of Motion 
Picture Arts & Sciences ; Metro-Gold wyn- Mayer Studies, Culver City, Calif. 




[J. S. M. P. E. 

maximum benefit from these improved recordings, for the theater 
reproducing equipment to advance in step with the progress of the 
recording art. 

Theater equipment that was considered adequate in the past is 
no longer capable of reproducing faithfully the current dialog, music, 
and sound effects now being recorded by the studios. Improvements 
in recording permit more faithful reproduction of the human voice as 
well as of vocal and instrumental music. Improvements in amplifiers 
permit a wider power range and allow an increased volume of sound 
on the film itself, and the theater reproducing apparatus must conse- 


FIG. 1. Recommended amplifier output in electric watts in 
terms of the floor area of the theater. 

quently be capable of transmitting this improved quality to the 
theater patrons. 

Recent developments in the reproducing equipment have included 
the introduction of horns of new design which give far better quality 
than was formerly possible and more even and adequate distribution 
of sound throughout the theater auditorium. Improvements in the 
film-running mechanism have reduced flutter to a minimum, and 
increased amplifier power is now available for reproducing ade- 
quately and without distortion the wider power ranges now being 
recorded on the film. 

It is recognized in the studios that until such time as all theaters 
are equipped with new and modern improved equipment, methods 
must be used that will allow the wider-power-range films to be repro- 
duced to their best advantage in those theaters having equipment 

Feb., 1938] 



capable of this reproduction, and will not, at the same time, penalize 
those theaters that are fitted with reproducing equipment not capable 
of handling the wider volume range. During the past year several 
of the major companies have, in a limited number of releases, made 
available to the theaters two general types of prints : one type being 
the "Regular" release print with the ordinary volume range, and the 
other type, divided into two classifications according to the volume 
range recorded on the film, known as "Hi-Range" and "Lo-Range" 

The "Hi-Range" prints, requiring increased amplifier power in the 
reproducing equipment, and having an approximate sound intensity 


20 3456 89100 


910000 20000 


FIG. 2. Recommended amplifier output in electric watts in 
terms of the volume of the theater. 

range of 50 db., produce intensity changes that closely approximate 
those occurring in nature. Musical passages so recorded and subse- 
quently reproduced with adequate power, lend the added color and 
naturalness necessary to insure more complete enjoyment of the 

Those productions released on "Hi-Range" prints will also be 
available on "Lo-Range" prints, the volume of which may correspond 
to the studio "Regular" prints, or may be recorded to play 3 or 4 
db. above the particular studio's average (Figs. 2 and 3). In other 
words, any production issued on "Regular" prints will be distributed 
completely on one type of print, while any production available 
on "Hi-Range" prints will also necessarily be available on "Lo- 
Range" prints as well. 

212 J. K. HlLLIARD [J. S. M. P. E. 

As more and more theaters are converted to the modern equipment 
capable of reproducing wider volume ranges, the practice of issuing 
"Hi-Range" and "Lo-Range" prints will undoubtedly be rapidly 

The success of such productions as May time, One Hundred Men and 
a Girl, and other similar musical productions released on "Hi-Range" 
prints, indicates that this type of release print has a definite place in 
the industry from a showmanship standpoint. Complete apprecia- 
tion by the exhibitor of the technic required for their reproduction 
will insure still greater box-office success. 

By means of improved technic in the studio, "Hi-Range" prints 
have a controlled balance of volume between dialog and music; 
that is, relative reproduction between the dialog and music has been 
predetermined by experienced showmen after careful consideration 
of the output level. 

The sound volume reaching the ear of a patron from any given 
print projected at a certain fader setting depends upon the percentage 
modulation of the signal on the film. On "Regular" prints (pro- 
jected at the average fader setting for any particular studio's product) , 
both the dialog and music are given 100 per cent modulation a 
greater part of the time. This means that the output volume will 
be practically the same throughout the production. 

In recording "Hi-Range" prints, however, most of the dialog 
passages are intentionally reduced in modulation so that the average 
dialog modulation rarely exceeds 50 per cent, while the music is 
recorded at 100 per cent modulation. This provides a volume dif- 
ferential between music and dialog of at least 6 db. "Hi-Range" 
prints do not necessarily provide louder sound, but an extended 
volume range that gives more dramatic value in the theater. 

When such a print is projected, the fader must be raised at least 6 db. 
for proper dialog volume. To utilize this volume range on the film 
the theater must necessarily be provided with an amplifier output 
that is increased by approximately the same range. 

Increased amplifier power is necessary since in the past the average 
theater installation has had only sufficient power to reproduce dialog 

In general, those theater installations equipped with modern loud 
speaker systems have sufficient amplifier power to reproduce ade- 
quately this higher volume range. 

By observation of a number of houses it has been found that a 

Feb., 1938] 



theater containing up to 1000 seats requires from 10 to 15 watts of 
power, from either the original old standard horn systems or the more 
modern two-way loud speaker systems. Houses having from 1000 
to 2000 seats require from 19 to 24 watts of power, and theaters with 
over 2000 seats require at least 48 watts. Houses equipped with the 
Electrical Research Products, Inc., three-way, wide-range system will 
require approximately the same power for the same seating capacity. 

In order to simplify the determination of power necessary for 
theaters of various sizes to reproduce adequately the greater volume 
ranges now being recorded charts have been prepared as follows : 

Fig. 1 shows the recommended amplifier power output in terms of 



FIG. 3. Recommended amplifier output in electric watts in 
terms of the seating capacity of the theater. 

the theater floor area; Fig. 2, in terms of the cubical contents of the 
theater; Fig. 3, in terms of the number of seats. 

These curves indicate the necessary amplifier capacity to maintain 
high quality of sound reproduction, but since the required power is de- 
pendent partially upon the absorption and reverberation characteris- 
tics of the theater auditorium, deviation from these values may be 
required depending upon the variation of any particular theater from 
optimal reverberation conditions. 

In reproducing a "Hi-Range" print, the theater manager and pro- 
jectionist should follow the usual method of setting the fader for 
proper dialog volume, which will automatically insure proper repro- 
duced volume level for any musical passages in the same production. 
If the volume level of the music is reduced to a point lower than that 


originally intended at the time of the recording, dialog passages 
would be too low for satisfactory reproduction. 

If the equipment is not functioning properly or if there is insuffi- 
cient power capacity, the higher-volume portions of the musical 
passages will be reproduced with harshness and distortion. In this 
type of reproduction, flutter (if present) due to poor motion of the 
film through the sound-gate will be particularly noticeable. When 
such prints are reproduced on the older types of theater systems the 
increase in amplification necessary to reproduce the high-volume 
passages properly will sometimes introduce objectionable hum and 
other system noises, which can usually be eliminated by careful ad- 
justment of the system. 

The use of the higher amplifier power necessary to reproduce these 
prints also requires that the distribution of sound throughout the 
theater be particularly uniform. 

In order to assist the exhibitors, theater managers, and projection- 
ists as well as the exchanges in identifying quickly the "Hi- Range" 
and "Lo-Range," as well as "Regular" prints, each of the major 
studios will commence immediately to label each print Hi-Range or 
Lo-Range, or Regular, and designate a general average fader setting 
at which the print should be projected this information to be in- 
cluded in the Standard Release Print Leader on each reel of each 
production. It is suggested that all theater projectionists carefully 
watch every print in order to take advantage oj this additional informa- 
tion which should assist in increasing the showmanship value of re- 
corded sound. 



Editorial Note: The following paper represents a section of the Technical Bulle- 
tin of the Research Council of the Academy of Motion Picture Arts & Sciences as 
published on November 24, 1937. It is reprinted in this Journal to give it wider 
distribution among members of the Society of Motion Picture Engineers. 

These Fader Setting Leader Instructions have been formally approved by the 
Academy Research Council but have not as yet received action by the Standards 
Committee of the Society of Motion Picture Engineers. 

To aid the exhibitor further in the proper handling of "Hi-Range" 
prints the studios will, commencing about December 1, 1937, utilize 
that part of the Academy Research Council Standard Release Print 
Leader that has been designated for use for any pertinent information 
to be transmitted from studio to theater. 

A portion of the specifications for the Standard Release Print 
Leader, indicating the location of this instructional information, is 
shown in Fig. 1, and details of the information to be known as Stand- 
ard Fader Setting Instructions are illustrated in Fig. 2. 


The Standard Fader Setting Instruction Leader shall consist of 15 
frames located as specified (Academy Research Council Standard Re- 
lease Print Leader) in the synchronizing leader; the first frame shall 
designate the type of print; the second frame the type of reproducing 
equipment necessary to project the print; and the next nine frames the 
general fader setting specified in relation to an average fader setting for 
the particular product under consideration. The remaining frames 
may be used for whatever additional information the studio may wish to 
transmit to the theater. 

This instruction leader will be of assistance to the exchanges in 
that it will facilitate the special handling required in the exchange for 
the various types of prints, by providing an easily noted means of 
identification for each type. 

It should be noted that the designation "Regular" in the Standard 
Fader Setting Instruction Leader indicates that only one type of 
print has been issued on the particular production under considera- 
tion. Productions with prints designated as either "Hi-Range" or 






Shall be either transparent or raw stock. 

When the protective leader has been reduced to a length of 

six feet it is to be restored to a length of eight feet. 


Shall contain 24 frames in each of which is plainly printed 
in black letters on white background: (a) type of print, 
(ft) reel number (Arabic numeral not less than l /4 of frame 
height), and (c) picture title. 


Shall consist of 20 frames ahead of Start mark, then 12 feet, 
including Start mark, to picture, opaque except as specified 
below: In the center of the first frame there shall be printed 
across the picture and sound-track area a white line l /3i 
inch wide upon which is superimposed a diamond 1 /s inch 

The next 15 frames may be used by the studio for sensito- 
metric or other information. If not so used this leader shal 
be opaque. 

The Start mark shall be the 21st frame, in which is printed 
START (inverted) in black letters on white background. 
The Academy camera aperture height of .631 inch shall be 
used in the photography of this frame, and all others between 
Start mark and beginning of picture. 

From the Start mark to the picture the leader shall contain 
frame lines which do not cross sound-track area. 

In the frames in which the numerals "6" and "9" appear, the 
words "six" and "nine" (also inverted) shall be placed imme- 
diately below the figure, to eliminate the possibility of mis- 
reading in the projection room due to the similarity between 
the inverted numerals. 

Beginning 3 feet from the first frame of picture, each foot is to 
be plainly marked by a transparent frame containing an in- 
verted black numeral at least l /z. frame in height. Footage 
indicator numerals shall run consecutively from 3 to 11 , inclu- 
sive. At a point exactly 20 frames ahead of the center of each 
footage numeral frame there shall be a diamond (white on 
black background) Vsinch high by */g inch wide. 

FIG. 1. 

Start of Picture 

For specifications for motor and change-over cue location and reel-end leader 
see complete Academy Research Council Specifications for 35-mm. Motion Pic- 
ture Release Prints in Standard 2000-ft. Lengths, published January 6, 1936. 










| PUSH PUtL j 





FIG. 2. Academy standard fader setting instructions. 


"Lo-Range" will have been issued in both types of print, i. e., all 
productions on "Hi-Range" prints necessarily will have been issued 
on "Lo-Range" prints as well. 

This instruction leader will also enable the projectionist to identify 
prints that require "push-pull" reproducing systems as contrasted to 
prints requiring "single" systems. 

In order to identify more plainly the "push-pull" or "single" sys- 
tem prints, it was decided to include both the terms "push-pull" and 
"single" on every leader, crossing out in the laboratory one or the 
other of these two to leave the appropriate term designating the type 
of sound-track on the print. Fig. 2 indicates the manner by 
which this was accomplished for leaders that would be included in 
prints containing sound-tracks for reproduction on a "single" system. 
For leaders to be included in prints containing "push-pull" tracks the 
word "single" would have been crossed out, leaving the word "push- 
pull" to indicate this type of track. 

In order that the exhibitor may achieve the best results, the fader 
setting designated in this leader should be followed in general, inas- 
much as the entire balance between the dialog and music throughout 
the reel will be chosen for each designated setting. 


During the Conventions oj the Society, symposiums on new motion picture appara- 
tus and materials are held, in which various manufacturers of equipment describe and 
demonstrate their new products and developments. Some of this equipment is de- 
scribed in the following pages; the remainder will be published in subsequent issues 
of the Journal. 



The portable unit described herein is one of seven such units recently built for 
Warner Bros. Pictures, Inc., by RCA Manufacturing Company, and is aptly 
termed "a mobile sound recording channel," for, in contrast to the average loca- 
tion truck, nothing of the performance of a fixed studio channel has been sacri- 
ficed in achieving portability. The equipment is of the highest grade and is in- 
terchangeable with similar units in use in the fixed channels. 

Fig. 1 shows the truck ready for recording, with the mixer case set up and the 
microphone connected. The body and cab are built on a standard four-speed, 
two-ton Ford chassis. The color scheme is red, blue, and white, with chromium 
strips to give the illusion of length. Actually, the wheel base is only 157 inches, 
and the overhang at the rear is small, so as to allow good maneuverability in 
woods and on mountain roads. All batteries are set low, resulting in a low cen- 
ter of gravity and good stability on curves and grades. The body is insulated 
with three inches of kapok, and the roof is painted with aluminum paint to reflect 
the heat of the sun. 

Instead of a microphone boom, a two-sectioned jointed duralumin pole is 
provided to hold the microphone. This pole can be held by hand or mounted 
upon a lamp-stand, and has proved far superior to other types of boom on loca- 
tion. The microphone is insulated from the pole by rubber. Wind and rain 
screens are provided for the microphone so that recordings can be made under all 
but the severest conditions. Running shots, disk and film playbacks, and process 
projection work can all be handled as easily with this unit as with any fixed studio 
installation, so that it has actually proved itself a mobile recording channel. 

Fig. 2 is a rear view, with the doors open. On the right are the large cable 
reels, operated by detachable handles from the inside of the truck. Sufficient 
cable is carried on the reels for recording at a distance of 1000 feet from the action. 
On the left is the plug panel for connections to the portable mixer and camera 
motors. All plugs are plainly marked and are poled so that it is impossible to 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received October 
4, 1937. 

** Warner Bros. Pictures, Inc., Burbank, Calif. 




FIG. 1. Location truck set up for recording. 

FIG. 2. Rear view of location truck. 

Feb., 1938] 





make mistakes in connections. Despite the low height of the truck, the head 
room is six feet one inch. The top of the door follows the curve of the body, 
eliminating stooping when entering. 

The body is divided into front and rear sections. Fig. 3 shows the interior of 
the rear compartment. On the left are the amplifier rack and the power control 
panel. In the center is the RCA ultraviolet light recording machine. Either 

FIG. 5. Front compartment. 

push-pull or standard bilateral shutter track may be made. The recording ma- 
chine connections are brought to plugs so that machines for different types of 
recording can readily be substituted. Beneath the recorder is space for ten 
double film magazines. The compartment can be converted easily into a dark- 
room for loading and unloading magazines on location if desired. 

The amplifier rack contains all the amplifiers, filters, and noise-reduction equip- 
ment necessary for recording. The rack is of duralumin, and the amplifier 


units are mounted on standard 19-inch aluminum panels finished in black alumi- 
lite. All input and output connections are brought out to jacks, to provide ease 
in testing and locating trouble. Each unit is individually fused in both high- and 
low-voltage circuits. A special airplane-type 250-volt dynamotor driven from 
the 110-volt motor battery supplies the amplifiers through appropriate filters, 
eliminating all "B" batteries. The noise-level for normal recording gain of the 
complete amplifier system, with all rotating equipment in operation, is 65 db. be- 
low 100 per cent modulation. This is considerably lower than the noise-level of 
many permanent installations. 

The power control panel is alongside the recorder and contains all controls in 
addition to those on the machine needed by the recordist during operation. At 
the top are the charge-discharge meters and the 110-volt d-c. voltmeter. Below 
these are the charge-discharge switches and the two end-cell switches for an addi- 
tional 6 or 12 volts when the service is unusually severe. In addition to the 110- 

FIG. 6. Mixer case. 

volt motor battery, there is an 8-volt amplifier battery and a 14-volt recording 
lamp battery. The truck can be used for two full days and nights of recording 
without recharging batteries, and the battery capacities are so chosen that all 
batteries are discharged in approximately the same time. Below the switches 
are the a-c. and d-c. voltmeters for interlock voltage and plate voltage and a large 
size speed indicator. Along the bottom of the panel are the starting switches and 
the lights and fan switches. Lights are of the tubular type, recessed in the ceiling 
on each side of the recorder, and an exhaust fan in the ceiling provides ventilation 
in hot weather. The lights and fan can be switched from the 110-volt battery to 
a 110-volt a-c. circuit if the latter is available from an outside source. 

Fig. 4 shows the amplifier rack swung out for servicing. The equipment covers 
have been removed so as to show the workmanlike job of wiring. The rack wiring 
is of the highest grade airplane type, No. 16 stranded wire, and is led to the rack 
in a length of flexible conduit. High- and low-level circuits are in separate forms 
on the rack and cross-talk is reduced to a negligible quantity. Only three types 
of standard RCA tubes are used throughout the entire recording system. 

The heavy wiring to the power control panel prevents swinging this panel out- 


ward, as in the case of the amplifier rack. The service door is located at the rear 
of the panel on the left side of the body. All parts are stencilled and all wiring 
is in conduit and is color-coded. At the top is a work light and at the base are 
soldering iron outlets. 

The front compartment of the truck, as shown in Fig. 5, contains all the rotat- 
ing equipment and the small accessories used on the set. By locating such equip- 
ment forward, the stage helper need not disturb the recordist as he collects his 
mixer case, telephone set, microphones, etc., and the recordist is left free to prepare 
his equipment for recording. The battery chargers are in the background. 
Auxiliary outlets are provided for the mixer case so that mixing can be done in the 
front compartment if desired. The windows can be lowered, and a ventilalitiK 

FIG. 7. Pre-amplifier and microphone. 

fan and lights are provided. The same type of selsyn motor system is provided 
as is used in the studio. This has been found to be very desirable and allows play- 
back and process projection equipment to be operated in synchronism with the 
cameras. The machine compartments are sound-proofed with lead and felt so 
that the truck is quiet enough to operate right on the set if desired. 

Fig. 6 shows the mixer case. It is built of duralumin with a black alumiliie 
finish, and is provided with a collapsible stand of the same material. Three mix- 
ing positions are available and facilities are provided to read and adjust heater 
and plate currents to the microphone pre-amplifiers. A high-speed extension 
volume indicator is supplemented by the latest type of high-quality monitoring 
head-phones. A telephone subset connects the mixer with the recordist and an 
extension connects with the stage helper at the microphone. 

Fig. 7 shows the portable pre-amplifier and the type 630 microphone. The 
pre-amplifier has two stages of amplification and a gain of 48 db. It can be 
plugged into the line anywhere between the microphone and the mixer case. On 

Feb., 1938] 



long microphone runs it is used at the microphone, to raise the speech level well 
above line noised causes by lighting cables or electrical interference of any kind 
on the set. A radio-frequency filter is incorporated in the circuit to prevent radio 
pick-up when recording near broadcast antennas and aboard ships. Fig. 8 shows 
the pre-amplifier with the cover removed. The tube shelf may be opened for ease 
of servicing. 

FIG. 8. Pre-amplifier, without case. 

Fig. 9 is a block diagram of the recording circuit. For recording music, the 
frequency response is flat within 1 db. from 40 to 9000 cps. For recording 
dialog the low end of the characteristic is reduced 6 db. at 100 cps. by a dialog 
equalizer in the pre-amplifier, the response is sharply cut off below 100 cps. by an 
80-cycle high-pass filter, and the high end is always cut off at 9000 cps. by the 
9000-cycle low-pass filter. Upward equalization is available by means of a high- 
frequency equalizer ahead of the recording amplifier. In general, a gradual 3-db. 
rise is used from 1000 to 7000 cps. 

The mixer amplifier and recording amplifier are two-stage amplifiers, each hav- 
ing a gain of 50 db. Two units are used for ease of inserting filters and to render 
the equipment interchangeable with fixed-channel equipment in which the mixer 
amplifier is in a tea-wagon console on the stage and the recording amplifier in a 
Central recording building. The overall harmonic distortion of the recording sys- 



tern is less than one per cent at an output of +22 db. above a zero power level of 
6 milliwatts. This output level is 6 db. above 100 per cent modulation of the re- 
cording galvanometer. 

The motor circuit consists of a 220-volt, 3-phase, 60-cycle, 1.75-kva. converter, 
driven by a 110-volt storage battery supplying interlock voltage to a 3-phase dis- 
tributor driven by a 1 /e-hp. 110-volt d-c. motor. As many as eight cameras can 
be interlocked with the recording machine and distributor. Normally the motor 

FIG. 9. Diagram of recording circuit. 

speed is held at 1200 rpm. by a flywheel and butterfly type of control on the d-c. 
motor. The control can be cut out, however, and rheostats used to vary the 
speed from 900 to 1400 rpm. for undercranking or overcranking the cameras. 
A synchronizing circuit is provided that fogs the film in the camera and opens 
the bias circuit of the recorder when the system is up to speed. A limited power 
supply at 110-volts a-c. for playback and public-address amplifiers is available 
from the converter through a 3-phase 220/1 10-volt transformer. 




Since the early days of silent pictures, the film measuring machine has been one 
of the tools of trade for the film cutter. The device here described is a modern- 
ized and improved form of the old familiar footage counter. Engineered to meet 
the exacting demands of musical conductor, re-recording mixer, commentators, 
and sound-effects men, it is a definite advance over previous devices. (Fig. 1.) 

Simplicity has been the keynote in its construction. It consists of a three-place 
Veeder counter driven by a synchronous clock motor assembled so as to create a 
minimum of noise, the clock-face being illuminated from inside the case and in- 

FIG. 1. Cueing device. 

clined so as to be easily read. The associated electric stop-watch has the advan- 
tage over the hand-watch in that it has a much larger dial and may be started 
either manually or by plugging in on the a-c. projector or recording circuit so that 
timing begins with the start-marks on the film, before the picture appears on the 
screen. Of course, this is true also of the footage counter. 

The case is of cast aluminum alloy and is so proportioned as to cause a 
minimum of resonance and sounding-board effect. The result is a compact and 
easily read device that may be used within a foot or two of the microphone, as is 
often necessary in connection with newsreel commentation. 

*Presented at the Fall, 1937, Meeting at New York, N. Y.; received October 
14, 1937. 
** Reeves Sound Studios, Inc., New York, N, Y, 


The applications of the counter seem to increase with use. Often the counters 
are used in pairs and sometimes even three's. Using the counters simultaneously 
saves many rehearsals and hundreds of feet of film. 

A counter with stop-clock is usually preferred in the control room. A counter 
without the clock is ordinarily sufficient for a commentator, and when scoring 
from high-quality disk recordings a third counter is pressed into service. 

A record or film may be started and run out to some predetermined length, say, 
39 feet. The projector and recording machine then are started at, let us say, 15 
feet. The mixer opens his controls, and if the cue sheet has been correctly pre- 
pared, the title music will begin exactly at the desired spot; and so on down 
through the reel. Often the commentator or the sound-effects must come in on 
a blind cue; the synchronized counter gives the correct instant for starting or 
stopping. Hands are left free, numbers are large and illuminated, and useless 
conversion from seconds to feet is eliminated. 

In the cutting and review rooms this convenience is also valuable. Unlike the 
regular stop-watches which have an unhandy way of getting lost at critical mo- 
ments, this is a piece of permanent equipment, always ready for instant use. Cor- 
rections, cuts, and inspections may be made directly in terms of feet from the be- 
ginning of the reel; or, when counter and clock combinations are used indepen- 
dently of each other, one may measure both the reel and the duration of a specific 

Simplicity, silence, and ruggedness are the important mechanical features of the 
device, and practical experience with the most exacting of cueing problems has 
been responsible for the present design. 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Library of the U. S. 
Department of Agriculture, Washington, D. C. 

American Cinematographer 

18 (Nov., 1937), No. 11 

Film Is Most Valuable in Television (p. 450). H. R. LUBCKE 

Ampro's Model L for All Houses (p. 473). 

18 (Dec., 1937), No. 12 
What 1937 Has Shown in Technical Progress in Motion 

Picture Making (p. 493). 

Soviet Working in New Stereoscopic Pictures (p. 502) . V. SOLVE v 
Walter Bell Completes 8-Mm. Reversal Machine 

(p. 519). 

American Physics Teacher 

5 (Dec., 1937), No. 6 

Advanced Laboratory Experiments in Acoustics, In- 
cluding a New Method for Measuring the Adsorption 
of Sound in Tubes (p. 252). C. K. STEDMAN 


17 (Nov., 1937), No. 11 
Disk Recording Record Processing (p. 24). T. L. DOWEY 

Educational Screen 

16 (Nov., 1937), No. 9 
Foreign Films for Educational Institutions (p. 289). M. Z. MERCIER 


10 (Nov., 1937), No. 11 
New Pictures by Wire (p. 12). 
Phonograph Pick-Up Tracking Error vs. Distortion and 

Record Wear (p. 19). B. OLNEY 

Screens for Television Tubes (p. 31). I. G. MALOFF AND 


International Photographer 

9 (Nov., 1937), No. 10 
Modern Backlot Magicians (p. 24). C. JONES 



New Continuous Projector (p. 27). 

The Soundman's Book of Tables (p. 20) J. N. A. HAWKINS 

The Laboratory Book of Tables (p. 22). D. K. ALLISON 

(The last two are the first of a series of tables, to be 
published consecutively.) 

International Projectionist 

12 (Nov., 1937), No. 11 
An Analysis of Imperfections Apparent on the Screen 

II (p. 7). A. C. SCHROEDER 

Fundamentals of Sound Recording and Theater Repro- 
duction IV (p. 11) F. T. JAMEY 

Typical Troubles in Modern Sound Reproducing Units 
F/(p.20). L. CHADBOURNE 

An Outline of Tube Types Used in Modern Sound- 
Picture Amplifiers (p. 24). W. STERLING 

New Service Supply Units to Operate in the Theater 

Field (p. 27). J. J. FINN 

Kinematograph Weekly 

249 (Nov. 18, 1937), No. 1596 

Gevaert in the Color Field in Association with True- 
Color (p. 52). 


19 (Oct., 1937), No. 11 

Die Reichsstelle fur den Unterrichtsfilm (Teaching 

Films in Germany) (p. 254). W. HELMBRECHT 

Dynamik und Reintonwirkung der verschiedenen 
Schriftsy steme in Tonfilm (Dynamics and Pure 
Tone Effects of Different Stylus Systems for Sound- 
Films) (p. 255). A. NARATH 

Der Belichtungsmesser des Kameramannes (The Cam- 
eraman's Exposure Meter) (p. 259). L. KUTZLEB 

Anforderungen an Gluhlampen fiir den Bildwurf (Re- 
quirements for Incandescent Lamps for Projection) 
(p. 261). O. HOPCKE 

Movie Makers 

12 (Dec., 1937), No. 12 

Robot 35-Mm. (p. 621). R. C. HOLSLAG 

Sound Booth (p. 634). 

Philips Technical Review 

2 (Aug., 1937), No. 8 
The Enlarged Projection of Television Pictures (p. 249). M. WOLF 

2 (Sept., 1937), No. 9 
The Relationship between Fortissimo and Pianissimo 

(p. 266). R. VERMEULEN 


Photographic Journal 

77 (Nov., 1937), No. 11 
The Special Effects Department (p. 607). H. CHEVALIER 

Photographische Industrie 

35 (Nov. 24, 1937), No. 47 

Fortschritte in der Fabrikation von Schmalfilmgeraten 
(Progress in the Manufacture of Substandard Film 
Apparatus) (p. 1269). 

Technique Cinematographique 

9 (Oct., 1937), No. 82 
Vers le cinema stereoscopique de 1'avenir (Motion 

Picture Stereoscopy of the Future) (p. 1019). V. SOLEF 


10 (Nov., 1937), No. 117 

A New Idea for Large-Screen Pictures (p. 653). 

The Design of the G.E.C. Television Receiver (p. 663). D. C. ESPLEY AND 





Officers and Committees in Charge 

W. C. KUNZMANN, Convention Vice-President 

J. I. CRABTREE, Editorial Vice-President 

G. E. MATTHEWS, Chairman, Papers Committee 

W. WHITMORE, Chairman, Publicity Committee 

E. R. GEIB, Chairman, Membership Committee 

N. D. GOLDEN, Chairman, Local Arrangements Committee 

Local Arrangements and Reception Committee 


N. D. GOLDEN, Chairman 





Registration and Information 

W. C. KUNZMANN, Chairman 



Ladies' Reception Committee 

MRS. R. EVANS, Hostess 

assisted by 



Banquet Committee 

R. EVANS, Chairman 



Publicity Committee 

W. WHITMORE, Chairman 




Convention Projection Committee 

H. GRIFFIN, Chairman 



Officers and Members of Washington Projectionist Local 224. 

Membership Committee 

E. R. GEIB, Chairman 


Hotel and Transportation Committee 

J. G. BRADLEY, Chairman 




The headquarters of the Convention will be the Wardman Park Hotel, where 
excellent accommodations are assured. A reception suite will be provided for the 
ladies, for whom also is to be arranged an interesting program of entertainment. 

By special arrangement with the Hotel Management, special breakfast, lunch- 
eon, and dinner service will be provided on the Continental Room Terrace, 
for SMPE delegates only. 

The following daily hotel rates, European plan, are guaranteed to SMPE 
delegates attending the Convention: 

One person, room and bath $ 3 . 50 

Two persons, standard bed 5.00 

Two persons, twin beds 6 . 00 

Parlor suite, one person 9 . 00 

Parlor suite, two persons 11 . 00 

Room reservation cards will be mailed to the membership of the Society in the 
near future, and those who plan to attend the Spring Convention should return 
their cards promptly to the Wardman Park Hotel to be assured satisfactory 
accommodations. Local railroad ticket agents should be consulted with regard to 
trains and rates. 

For those who will motor to the Convention ample free parking space is avail- 

234 SPRING CONVENTION [j. s. M. p. E. 

able on the Hotel grounds. For those who prefer parking in the Hotel garage, 
a special rate of 75 cents a day has been arranged. 

Technical Sessions 

An attractive and interesting program of technical papers is being assembled 
by the Papers Committee. All technical sessions, apparatus symposiums, and 
film programs will be held in the Little Theatre of the Hotel. 

Apparatus Exhibit 

An exhibit of newly developed motion picture apparatus will be held, to which 
all manufacturers of equipment are invited to contribute. No charge will be 
made for space. Information concerning the exhibit and reservations for space 
should be made by writing to the General Office of the Society. 

Apparatus displayed should be newly designed or developed, or should have 
features of technical interest for the engineers attending the Convention. 

Registration and Information 

The Convention registration headquarters will be located at the entrance of the 
Little Theatre, where all the technical sessions will be held. The members of the 
Society and guests attending the Convention are expected to register and receive 
their badges and identification cards for admittance to special evening sessions. 
These cards will also be honored at several de luxe motion picture theaters in 
Washington during the four days of the Convention. 

Informal Luncheon and Semi-Annual Banquet 

The usual informal Luncheon will be held at noon of the opening day of the 
Convention, April 25th, in the Continental Room of the Hotel. On the evening 
of Wednesday, April 27th, will be held the Semi-Annual Banquet of the Society 
in the Continental Room of the Hotel at 8 :00 P.M. Addresses will be delivered 
by prominent members of the industry, followed by dancing and other entertain- 

Motion Pictures 

Delegates registering at the Convention will be supplied with complimentary 
passes to the following motion picture theaters in Washington during the dates of 
the Convention: 

By courtesy of Mr. J. J. Payette: Warners' Uptown and Earl Theaters. 

By courtesy of Mr. H. Meiken: RKO Keith's Theater. 

By courtesy of Mr. C. Barron: Loew's Capitol, Palace, and Columbia Theaters. 

Feb., 1938] SPRING CONVENTION 235 

Points of Interest 

To list all the points of interest in and about Washington would require too 
much space, but among them may be mentioned the various governmental 
buildings, such as the Capitol, the White House, Library of Congress, Depart- 
ment of Commerce, U. S. Treasury, U. S. Bureau of Standards, Department of 
Justice, Archives Building; and other institutions such as the National Academy 
of Sciences, the Smithsonian Institution, George Washington University, Wash- 
ington Cathedral, Georgetown University, etc. In addition may be included the 
Lincoln Memorial, the Washington Monument, Rock Creek Park, The Francis 
Scott Key Memorial Bridge, Arlington Memorial Bridge, the Potomac River, 
and Tidal Basin. Mt. Vernon, birthplace of Washington, is but a short distance 
away and many other side trips may be made conveniently via the many highways 
radiating from Washington. 


The Wardman Park Hotel management is arranging for golfing privileges for 
SMPE delegates at several courses in the neighborhood. Regulation tennis 
courts are located upon the Hotel property, and riding stables are within a short 
distance of the Hotel. Trips may be arranged to the many points of interest in 
and about Washington. 



The regular January meeting of the Board of Governors was held at the Hotel 
Pennsylvania, New York, N. Y., on January 14th. New members present at the 
meeting were E. A. Williford, Financial Vice-President, and R. E. Farnham, 
Governor. (A complete list of members of the Board of Governors will be found 
on the reverse of the Contents page of this issue.) 

Messrs. W. C. Kunzmann and J. I. Crabtree reported on preliminary plans for 
the Washington Convention, to be held April 25th-28th, inclusive, at the Ward- 
man Park Hotel. Some details of the Convention are given in the preceding sec- 
tion of this issue of the JOURNAL, and a complete program will be mailed to the 
membership in the near future. Detroit was selected for the Fall, 1938, Conven- 

Mr. O. M. Glunt, retiring Financial Vice-President, reported very satisfactory 
fiscal conditions for the year 1937, and in presenting the budget for 1938, Mr. E. A. 
Williford, Financial Vice-President elect, prognosticated an equally successful year 
for 1938. The membership is continuing to grow, having reached an all-time high 
of nearly 1300 members. 

The remainder of the Board meeting was concerned mainly with routine mat- 
ters. The next meeting will be held at Washington, D. C., April 24th, the day 
preceding the opening of the Spring Convention. 


On January 18th, the Mid- West Section held its regular monthly meeting in the 
Engineering Building, Chicago, 111., at which time Mr. A. F. Conto, formerly 
Chief Engineer of the Western Television Corp., presented a paper on the subject 
of "Trends in Television." The program was long and interesting, and the 
meeting was very well attended. The next meeting of the Section will be held on 
Tuesday, February 15th. 


At a meeting held at the Office of the Society on January 7th, the work of review- 
ing the present standards of the Society was completed. The material is being 
set in type and proofs are being mailed to all the members of the Standards Com- 
mittee, together with ballots for voting upon adoption of the revision. 

The revision is based upon the previous edition of the Standards, published in 
November, 1934, and contains all the changes and additions that have evolved 
in the interim as a result of developments in the motion picture art. A few new 
drawings have been added in the 35-mm. and 16-mm. categories, as also a com- 
plete set of 8-mm. drawings. Quite a number of tolerances have been added that 
were lacking in the previous edition. 


It is hoped that the voting upon the revision by the Standards Committee will 
be completed in time to publish the entire body of standards in the March issue 
of the JOURNAL, at which time comments from the membership of the Society will 
be solicited. If no objections arise, the Board of Governors will take action at 
their next meeting (April 24th) on adopting them as SMPE standards, after which 
they will be submitted to the Sectional Committee on Motion Pictures (ASA) for 
approval by the American Standards Association. Although a number of these 
standards have already been approved by the ASA, the complete body of standards 
will be re-submitted in view of the new plan of numbering and the single-sheet 


At a meeting held at the Paramount Building, New York, N. Y., on January 
20th, an agenda was established for the new year, and work continued on matters 
left unfinished from the previous year. Analysis of the theater survey charts is 
progressing, and it is anticipated that a report on the subject will be forthcoming 
at the approaching Washington Convention. 



Article I 


The name of this association shall be SOCIETY OF MOTION PICTURE 

Article II 


Its objects shall be : Advancement in the theory and practice of motion picture 
engineering and the allied arts and sciences, the standardization of the equipment, 
mechanisms, and practices employed therein, the maintenance of a high profes- 
sional standing among its members, and the dissemination of scientific knowledge 
by publication. 

Article III 


Any person of good character may be a member in any class for which he is 

Article IV 


The officers of the Society shall be a President, a Past-President, an Executive 
Vice-President, an Engineering Vice-President, an Editorial Vice-President, a 
Financial Vice-President, a Convention Vice-President, a Secretary, and a Trea- 

The term of office of the President and Past-President shall be two years; of 
the Engineering, Editorial, Financial, and Convention Vice-Presidents, two years; 
and of the Executive Vice-President, Secretary, and Treasurer, one year. Of the 
Engineering, Editorial, Financial, and Convention Vice-Presidents, two shall be 
elected alternately each year or until their successors are chosen. The President 
shall not be immediately eligible to succeed himself in office. 

Article V 

Board of Governors 

The Board of Governors shall consist of the President, the Past-President, the 
five Vice-Presidents, the Secretary, the Treasurer, the Section Chairmen, and five 
elected Governors. Two, and three, of the Governors shall be elected alternately 
each year to serve for two years. 

* Corrected to January 1, 1938. 


Article VI 


There shall be an annual meeting, and such other meetings as stated in the By- 

Article VII 

This Constitution may be amended as follows : Amendments shall be approved 
by the Board of Governors, and shall be submitted for discussion at any regular 
members' meeting. The proposed amendment and complete discussion then shall 
be submitted to the entire Active, Fellow, and Honorary membership, together 
with letter ballot as soon as possible after the meeting. Two-thirds of the vote 
cast within sixty days after mailing shall be required to carry the amendment. 

By-Law I 


Sec. 1. The membership of the Society shall consist of Honorary members, 
Fellows, Active members, Associate members, and Sustaining members. 

An Honorary member is one who has performed eminent services in the ad- 
vancement of motion picture engineering or in the allied arts. An Honorary 
member shall be entitled to vote and to hold any office in the Society. 

A Fellow is one who shall not be less than thirty years of age and who shall 
comply with the requirements of either (a) or (6) for Active members and, in 
addition, shall by his proficiency and contributions have attained to an outstand- 
ing rank among engineers or executives of the motion picture industry. A 
Fellow shall be entitled to vote and to hold any office in the Society. 

An Active member is one who shall be not less than 25 years of age, and shall be : 

(a) A motion picture engineer by profession. He shall have been engaged 
in the practice of his profession for a period of at least three years, and shall have 
taken responsibility for the design, installation, or operation of systems or appa- 
ratus pertaining to the motion picture industry. 

(6) A person regularly employed in motion picture or closely allied work, 
who by his inventions or proficiency in motion picture science or as an executive 
of a motion picture enterprise of large scope, has attained to a recognized standing 
in the motion picture industry. In case of such an executive, the applicant must 
be qualified to take full charge of the broader features of motion picture engi- 
neering involved in the work under his direction. 

(c) An Active member is privileged to vote and to hold sectional office, and 
to have full privileges in all activities of the Society except that he may not be 
elected to a national office. 

An Associate member is one who shall be not less than 18 years of age, and 
shall be a person who is interested in or connected with the study of motion 
picture technical problems or the application of them. An Associate member is 
not privileged to vote, to hold office or to act as chairman of any committee, 
although he may serve upon any committee to which he may be appointed; and, 
when so appointed, shall be entitled to the full voting privileges of a committee 


A Sustaining member is an individual, a firm, or corporation contributing 
substantially to the financial support of the Society. 

Sec. 2. All applications for membership or transfer shall be made on blank 
forms provided for the purpose, shall give a complete record of the applicant's 
education and experience. 

Sec. 3. (a) An Honorary membership may be granted upon recommendation 
of the Board of Governors when confirmed by a four-fifths majority vote of the 
Honorary members, Fellows, and Active members present at any regular meeting 
of the Society. An Honorary member shall be exempt from all dues. 

(&) Applicants for the grade of Fellow shall give as reference at least three 
Fellows in good standing. Applicants shall be elected to membership by the 
approval of at least three-fourths of the Board of Governors. 

(c) Applicants for Active membership shall give as reference at least three 
members of Active or of higher grade in good standing. Applicants shall be 
elected to membership by the approval of at least three-fourths of the Board of 

(d) Applicants for Associate membership shall give as reference at least one 
member of higher grade in good standing. Applicants shall be elected to member- 
ship by the approval of at least three-fourths of the Board of Governors. 

By-Law II 


Sec. 1. An officer or governor shall be an Honorary member, a Fellow, or an 
Active member. After January 1, 1935, Active members shall not be eligible to 
hold national office in the Society. 

Sec. 2. Vacancies in the Board of Governors shall be filled by the Board of 
Governors until the annual meeting of the Society. 

By-Law III 

Board of Governors 

Sec. 1. The Board of Governors shall transact the business of the Society be- 
tween members' meetings, and shall meet at the call of the president. 

Sec. 2. A majority of the Board of Governors shall constitute a quorum at 
regular meetings. 

Sec. 3. When voting by letter ballot, a majority affirmative vote of the total 
membership of the Board of Governors shall carry approval, except as otherwise 

Sec. 4. The Board of Governors, when making nominations to office, and 
to the Board, shall endeavor to nominate persons, who in the aggregate are 
representative of the various branches or organizations of the motion picture in- 
dustry, to the end that there shall be no substantial predominance upon the Board, 
as the result of its own action, of representatives of any one or more branches or 
organizations of the industry. 

By-Law IV 


Sec. 1. The location of each meeting of the Society shall be determined by the 
Board of Governors 


Sec. 2. Only Honorary members, Fellows, and Active members shall be en- 
titled to vote. 

Sec. 3. A quorum of the Society shall consist in number of one-tenth of the 
total number of Honorary members, Fellows, and Active members as listed in the 
Society's records at the close of the last fiscal year. 

. Sec. 4. The fall convention shall be the annual meeting. 

Sec. 5. Special meetings may be called by the president and upon the request 
of any three members of the Board of Governors not including the president. 

Sec. 6. All members of the Society in any grade shall have the privilege of dis- 
cussing technical material presented before the Society or its Sections. 

By-Law V 

Duties of Officers 

Sec. 1. The president shall preside at all business meetings of the Society and 
shall perform the duties pertaining to that office. As such he shall be the chief 
executive of the Society, to whom all other officers shall report. 

Sec. 2. In the absence of the president, the officer next in order as listed in 
Article 4 of the Constitution shall preside at meetings and perform the duties of 
the president. 

Sec. 3. The five vice-presidents shall perform the duties separately enumerated 
below for each office, or as defined by the president: 

(a) The executive vice-president shall represent the president in such geo- 
graphical areas of the United States as shall be determined by the Board of 
Governors, and shall be responsible for the supervision of the general affairs of 
the Society in such areas, as directed by the president of the Society. 

(b) The engineering vice-president shall appoint all technical committees. 
He shall be responsible for the general initiation, supervision, and coordination of 
the work in and among these committees. He may act as chairman of any com- 
mittee or otherwise be a member ex-officio. 

(c) The editorial vice-president shall be responsible for the publication of 
the Society's JOURNAL and all other technical publications. He shall pass upon 
the suitability of the material for publication, and shall cause material suitable 
for publication to be solicited as may be needed. He shall appoint a papers 
committee and an editorial committee. He may act as chairman of any com- 
mittee or otherwise be a member ex-officio. 

(d) The financial vice-president shall be responsible for the financial opera- 
tions of the Society, and shall conduct them in accordance with budgets approved 
by the Board of Governors. He shall study the costs of operation and the in- 
come possibilities to the end that the greatest service may be rendered to the 
members of the Society within the available funds. He shall submit proposed 
budgets to the Board. He shall appoint at his discretion a ways and means 
committee, a membership committee, a commercial advertising committee, and 
such other committees within the scope of his work as may be needed. He may 
act as chairman of any of these committees or otherwise be a member ex-officio. 

(e) The convention vice-president shall be responsible for the national con- 
ventions of the Society. He shall appoint a convention arrangements com- 


mittee, an apparatus exhibit committee, and a publicity committee. He may 
act as chairman of any committee, or otherwise be a member ex-officio. 

Sec. 4. The secretary shall keep a record of all meetings; he shall conduct the 
correspondence relating to his office, and shall have the care and custody of 
records, and the seal of the Society. 

Sec. 5. The treasurer shall have charge of the funds of the Society and disburse 
them as and when authorized by the financial vice-president. He shall make 
an annual report, duly audited, to the Society, and a report at such other times 
as may be requested. He shall be bonded in an amount to be determined by the 
Board of Governors and his bond filed with the secretary. 

Sec. 6. Each officer of the Society, upon the expiration of his term of office, 
shall transmit to his successor a memorandum outlining the duties and policies 
of his office. 

By-Law VI 


Sec. 1. (a) All officers and five governors shall be elected to their respective 
offices by a majority of ballots cast by the Active, Fellow, and Honorary members 
in the following manner: 

Not less than three months prior to the annual fall convention, the Board of 
Governors, having invited nominations from the Active, Fellow, and Honorary 
membership by letter form not less than forty days before the Board of Governors' 
meeting, shall nominate for each vacancy several suitable candidates. The sec- 
retary shall then notify these candidates of their nomination, in order of nomina- 
tion, and request their consent to run for office. From the list of acceptances, 
not more than two names for each vacancy shall be selected by the Board of 
Governors and placed on a letter ballot. A blank space shall also be provided 
on this letter ballot under each office, in which space the names of any Fellows or 
Honorary members other than those suggested by the Board of Governors may 
be voted for. The balloting shall then take place. 

The ballot shall be enclosed in a blank envelope which is enclosed in an outer 
envelope bearing the secretary's address and a space for the member's name and 
address. One of these shall be mailed to each Active, Fellow, and Honorary 
member of the Society, not less than forty days in advance of the annual fall con- 

The voter shall then indicate on the ballot one choice for each office, seal the 
ballot in the blank envelope, place this in the envelope addressed to the secretary, 
sign his name and address on the latter, and mail it in accordance with the in- 
structions printed on the ballot. No marks of any kind except those above pre- 
scribed shall be placed upon the ballots or envelopes. 

The sealed envelope shall be delivered by the secretary to a committee of tell- 
ers appointed by the president at the annual fall convention. This committee 
shall then examine the return envelopes, open and count the ballots, and announce 
the results of the election. 

The newly elected officers and governors of the general Society shall take office 
on the January 1st following their election. 

(6) The first group of vice-presidents, viz., the executive vice-president, engi- 
neering vice-president, editorial vice-president, financial vice-president, conven- 


tion vice-president, and a fifth governor, shall be nominated by the Board of 
Governors at its first meeting after the ratification of the corresponding provisions 
of the Constitution; and the membership shall vote on the candidates in accord- 
ance with the procedure prescribed in these By-Laws for regular elections of 
officers so far as these may be applicable. The term of these vice-presidents shall 
be deemed to begin January 1, 1934. 

By-Law VII 

Dues and Indebtedness 

Sec.l. The annual dues shall be twenty dollars ($20) for Fellows, ten dollars 
($10) for Active members, and six dollars ($6) for Associate members, payable on 
or before January 1st of each year. Current or first year's dues for new members, 
dating from the notification of acceptance in the Society, shall be prorated on a 
monthly basis. Five dollars of these dues shall apply for annual subscription to 
the publication. No admission fee will be required in any grade of membership. 

Sec. 2. (a) Transfer of membership may be made effective at any time by 
payment of the pro rata dues for the current year. 

(&) No credit shall be given for annual dues in a membership transfer from a 
higher to a lower grade, and such transfers shall take place on January 1st of each 

(c) The Board of Governors upon their own initiative and without a transfer 
application may elect, by the approval of at least three-fourths of the Board, any 
Associate or Active member for transfer to any higher grade of membership. 

Sec. 3. Annual dues shall be paid in advance. All Honorary Members, Fel- 
lows, and Active Members in good standing, as defined in Sec. 5, may vote or 
otherwise participate in the meetings. 

Sec. 4. Members shall be considered delinquent whose dues remain unpaid for 
four months. Members who are in arrears of dues for 30 days after notice of such 
delinquency, mailed to their last address of record, shall have their names posted 
at the Society's headquarters, which shall be the General Office, and notices of 
such action mailed to them. Two months after becoming delinquent, members 
shall be dropped from the rolls if non-payment is continued. 

Sec. 5. Any member may be suspended or expelled for cause by a majority 
vote of the entire Board of Governors; provided he shall be given notice and a 
copy in writing of the charges preferred against him, and shall be afforded oppor- 
tunity to be heard ten days prior to such action. 

Sec. 6. The provisions of Section 1 to 4, inclusive, of this By-Law VII, given 
above may be modified or rescinded by action of the Board of Governors. 

By-Law VIII 


Sec. 1. The emblem of the Society shall be a facsimile of a four-hole film -reel, 
with the letter S in the upper center opening, and the letters M, P, and E, in the 
three lower openings, respectively. In the printed emblem, the four-hole open- 
ings shall be orange, and the letters black, the remainder of the insignia being black 
and white. The Society's emblem may be worn by members only. 


By-Law IX 


Sec. 1. Papers read at meetings or submitted at other times, and all material 
of general interest shall be submitted to the editorial board, and those deemed 
worthy of permanent record shall be printed in the JOURNAL. A copy of each 
issue shall be mailed to each member in good standing to his last address of record. 
Extra copies of the JOURNAL shall be printed for general distribution and may be 
obtained from the General Office on payment of a fee fixed by the Board of 

By-Law X 

Local Sections 

Sec. 1 . Sections of the Society may be authorized in any state or locality where 
the Active, Fellow, and Honorary membership exceeds 20. The geographic 
boundaries of each Section shall be determined by the Board of Governors. 

Upon written petition, signed by 20 or more Active members, Fellows, and 
Honorary members, for the authorization of a Section of the Society, the Board of 
Governors may grant such authorization. 


Sec. 2. All members of the Society of Motion Picture Engineers in good stand- 
ing residing in that portion of any country set apart by the Board of Governors 
tributary to any local Section shall be eligible for membership in that Section, and 
when so enrolled they shall be entitled to all privileges that such local Section 
may, under the General Society's Constitution and By-Laws, provide. 

Any member of the Society in good standing shall be eligible for non-resident 
affiliated membership of any Section under conditions and obligations prescribed 
for the Section. An affiliated member shall receive all notices and publications 
of the Section but he shall not be entitled to vote at Sectional meetings. 

Sec. 3. Should the enrolled Active, Fellow, and Honorary membership of a 
Section fall below 20, or should the technical quality of the presented papers fall 
below an acceptable level, or the average attendance at meetings not warrant the 
expense of maintaining the organization, the Board of Governors may cancel its 


Sec. 4. Each Section shall nominate and elect a chairman, two managers, and 
a secretary-treasurer. The Section chairmen shall automatically become mem- 
bers of the Board of Governors of the General Society, and continue in that posi- 
tion for the duration of their terms as chairmen of the local Sections. 


Sec, 5. The officers of a Section shall be Active, Fellow, or Honorary members 
of the General Society. They shall be nominated and elected to sectional office 
under the method prescribed under By-Law VI, Section 1, for the nomination 
and election of officers of the General Society. The word manager shall be sub- 
stituted for the word governor. All Section officers shall hold office for one year, 


or until their successors are chosen, except the Board of Managers, as hereinafter 


Sec. 6. The Board of Managers shall consist of the Section chairman, the Sec- 
tion past-chairman, the Section secretary-treasurer, and two Active, Fellow, or 
Honorary members, one of which last named shall be elected for a two-year term, 
and one for one year, and then one for two years each year thereafter. At the 
discretion of the Board of Governors, and with their written approval, this list 
of officers may be extended. 

Sec. 7. The business of a Section shall be conducted by the Board of Managers, 


Sec. 8. (a) As early as possible in the fiscal year, the secretary of each Section 
shall submit to the Board of Governors of the Society a budget of expenses for the 

(b) The treasurer of the General Society may deposit with each Section secre- 
tary-treasurer a sum of money, the amount to be fixed by the Board of Governors, 
for current expenses. 

(c) The secretary-treasurer of each Section shall send to the treasurer of the 
General Society, quarterly or on demand, an itemized account of all expenditures 
incurred during the preceding interval. 

(d) Expenses other than those enumerated in the budget, as approved by the 
Board of Governors of the General Society, shall not be payable from the general 
funds of the Society without express permission from the Board of Governors. 

(e) A Section Board of Managers shall defray all expenses of the Section not 
provided for by the Board of Governors, from funds raised locally by donation, or 
by fixed annual dues, or by both. 

(/) The secretary of the Society shall, unless otherwise arranged, supply to 
each Section all stationery and printing necessary for the conduct of its business. 


Sec. 9. The regular meetings of a Section shall be held in such places and at 
such hours as the Board of Managers may designate. 

The secretary-treasurer of each Section shall forward to the secretary of the 
General Society, not later than five days after a meeting of a Section, a statement 
of the attendance and of the business transacted. 


Sec. 10. Papers shall be approved by the Section's papers committee previ- 
ously to their being presented before a Section. Manuscripts of papers presented 
before a Section, together with a report of the discussions and the proceedings of 
the Section meetings, shall be forwarded promptly by the Section secretary- 
treasurer to the secretary of the General Society. Such material may, at the dis- 
cretion of the board of editors of the General Society, be printed in the Society's 



Sec. 11. Sections shall abide by the Constitution and By -Laws of the Society, 
and conform to the regulations of the Board of Governors. The conduct of Sec- 
tions shall always be in conformity with the general policy of the Society as fixed 
by the Board of Governors. 

By-Law XI 


Sec. 1. These By-Laws may be amended at any regular meeting of the Society 
by a two-thirds vote by ballot of the members present at the meeting, a quorum 
being present, either on the recommendation of the Board of Governors or by a 
recommendation of the Board of Governors signed by any ten members of Active 
or higher grade. 




Volume XXX MARCH, 1938 Numbers 



Revision of SMPE Standards Proposed for Adoption by the 

Society 249 

Report of the Standards Committee 292 

Report of the Studio Lighting Committee 294 

Report of the Committee on Preservation of Film 300 

Changing Aspects of the Film-Storage Problem . J. G. BRADLEY 303 

The Practice of Projection A. N. GOLDSMITH 318 

Grading Projectionists G. P. BARBER 320 

Cooperation as the Keynote of Projection Service 

T. P. HOVER 326 

A Discussion of Screen-Image Dimensions. .F. H. RICHARDSON 334 

Perforated Screens and Their Faults F. H. RICHARDSON 339 

Commercial 16-Mm. Projection Faults C. L. GREENE 342 

Careless Work in Printing I. GORDON 347 

Current Literature 352 

Spring Convention at Washington, D. C., April 25-28, 1938.. . 354 

Society Announcements 358 





Board of Editors 
J. I. CRABTREE, Chairman 



Subscription to non-members, $8.00 per annum ; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West-Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1938, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers hi question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 


"President: S. K. WOLF, RKO Building, Rockefeller Center, New York, N. Y. 
"Past-President: H. G. TASKER, Universal City, Calif. 
"Executive Vice-President: K. F. MORGAN, 6601 Romaine St., Los Angeles, 


""Engineering Vice-President: L. A. JONES, Kodak Park, Rochester, N. Y. 
"Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 
""Financial Vice-President: E. A. WILLIFORD, 30 E. 42nd St., New York, N. Y. 
"Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 
"Secretary: J. FRANK, JR., 90 Gold St., New York, N. Y. 
"Treasurer: L. W. DAVEE, 250 W. 57th St., New York, N. Y. 


*J. O. AALBERG, 157 S. Martel St., Los Angeles, Calif. 
*M. C. BATSEL, Front and Market Sts., Camden, N. J. 
**R. E. FARNHAM, Nela Park, Cleveland, Ohio. 
*G. FRIEDL, JR., 90 Gold St., New York N. Y. 
*A. N. GOLDSMITH, 444 Madison Ave., New York N. Y. 
**H. GRIFFIN, 90 Gold St., New York, N. Y. 

**A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 
*S. A. LUKES, 6145 Glenwood Ave., Chicago, 111. 
*Term expires December 31, 1938. 
**Term expires December 31, 1939. 


The following edition of the Standards of the Society of Motion Picture Engineers 
is based upon the previous edition, published in November, 1934, and contains all the 
revisions and additions that have evolved in the interim as a result of developments in 
the motion picture art. A few new drawings have been added in the 35-mm. and 
16-mm. categories, as also a complete set of 8-mm. drawings. Quite a number of 
tolerances have been added, where lacking in the previous edition. 

The complete body of standards herein contained have received initial approval by 
the Standards Committee, and comments from the membership of the Society are 
solicited. If no objections arise, the Board of Governors will take action at their next 
meeting on adopting them as SMPE standards, after which they will be submitted 
to the Sectional Committee on Motion Pictures (ASA) for approval by the American 
Standards Association. Although a number of these standards have already been 
approved by the ASA, the complete body of standards will be re-submitted in view of 
the new plan of numbering and the single-sheet format. 

The early standardization activities of the Society have been de- 
scribed in an article entitled "A Historical Summary of Standardiza- 
tion in the Society of Motion Picture Engineers," published in the 
October, 1933, issue of the JOURNAL. Various editions of the standards 
promulgated by the Society have been published from time to time, 
the more comprehensive editions occurring in 1928, 1930, 1932, and 
1934. In addition to publication in the JOURNAL, the standards were 
printed in the form of small booklets entitled Dimensional Standards 
for Motion Picture Apparatus, and Recommended Practice. 

The issue published in November, 1934, was a revision of the edi- 
tion published in May, 1930, and approved by the American Stand- 
ards Association, September 20, 1930. Although some of the charts 
in the 1930 edition were superseded in the 1934 edition, the superseded 
charts were retained for purposes of reference, the original chart 
numbers remaining unchanged. Some of the changes involved new 
dimensions, whereas others involved merely a newer and clearer pres- 
entation of the data. The 1934 edition, therefore, contained all the 
15 charts previously published in 1930, in addition to new charts, 16 
to 32, inclusive. 

During the period folio wing the 1934 edition, considerable standardi- 

* NOTE: Comments on any or all the subjects presented in this report are 
invited from the membership of the Society. 



zation activity occurred throughout the world, particularly with re- 
spect to 16-mm. film and equipment. As a result of this activity, and 
also as a result of continuing development of the art, a need for sub- 
stantial revision of many of the standards drawings was indicated, 
not only for the purpose of keeping step with the developments, but 
also for the purposes of clarifying the material and presenting the 
dimensional data in such fashion as to avoid all possibility of misinter- 
pretation, to enhance their value in all other respects, and to devise 
a format that would admit of closer agreement with European stand- 

The present edition, therefore, is based upon the edition of 1934. 
Many of the charts have been redrawn, although for the most part 
the factual material has not been changed. The most important di- 
mensional changes occur in the drawings of the sound-tracks, as is to 
be expected in view of the rapid changes being made in the art of 
sound recording. A number of new drawings have been added, refer- 
ring particularly to those indicating the relation between the location 
of the sound-track, the emulsion surface of the film, and the direction 
of travel of the film. Also, tolerances have been added, where missing 
in the previous edition. However, no tolerances have been assigned 
to difference-dimensions, which in all cases have been calculated from 
the principal dimensions without considering tolerances. 

In view of the Society's affiliation with the American Standards 
Association and the International Standards Association, and, accord- 
ing with the general practice in international standardization, the 
basic dimensions in the present revision of the standards have been 
referred to the metric system, the English equivalents being calcu- 
lated to the appropriate number of decimal places. By so doing, the 
way is paved for much closer agreement in international standardiza- 

Up to the present time, all the standards adopted by the Society 
and approved by the American Standards Association have been ap- 
proved by the latter organization as a unit body of standards, to 
which has been assigned the designation ASA-Z22. In view of the 
inflexibility of the arrangement, the present purpose is to submit 
the standards for adoption by the ASA in the form of single-sheet 
standards to each of which designation numbers will be assigned 
individually, thus permitting revision of any drawing or chart and its 
re-submission to the ASA without affecting the designations of the 
other charts. The designations will be numbered consecutively as 


approved by the ASA, and will, in addition, retain the project designa- 
tion Z22. Thus, Z22.12 refers to the twelfth standard in the Z22 
project of the ASA, which is the motion picture project. 

In addition, however, an SMPE numbering scheme has been 
devised in order to include material that may not possibly be subject 
or amenable to national standardization through the ASA, and to 
supply a means of identification of SMPE standards prior to adoption 
by the ASA and assignment of designation numbers. Accordingly 
four categories have been established, referring principally to the 
widths of film in current use; and the letter m, referring to items not 
falling in those categories : 


16-mm. (doubly perforated) 
16-mm. (singly perforated) 
m miscellaneous 

The width designation is followed by the subject number in that 
category, which is again followed by the number of that particular 
drawing. The entire designation is preceded by the letters DS or RP, 
referring to Dimensional Standards or Recommended Practice, re- 
spectively. Thus DS35-5-1 means the first drawing (not revision: 
the first revision would be the second drawing) of Dimensional Stand- 
ard No. 5 in the 35-mm. category. In view of the fact that the num- 
ber of revisions made prior to this edition of the standards is uncer- 
tain, all the drawings in this edition are designated 1, so that accurate 
track of future revisions may henceforth be kept. As a matter of 
record, however, the original dates of the standards, as far as they 
could be traced, are included on the drawings. 

The entire standardization project is outlined in Table I, in which 
are shown the ASA "Z" numbers. 

The following sections list the dimensional standards and recom- 
mended practices. Where changes from the previous standards have 
been made, these changes have been indicated. 

35-Mm. (DS35) 

DS35-1-1 (Z22.1) Cutting and Perforating Negative and Positive 
Raw Stock. {1934 Chart 16) In dimensions C and D, the tolerances 
have been slightly widened. 



. S. M. P. E. 

Perforations. The perforation shown is the standard SMPE 
perforation, adopted July 14, 1933, for both positive and negative 

DS35-2-1 (Z22.2) Sprockets. (1934 Chart 19) Revised drawing. 
No changes in data except rounding off decimal places. 

DS35-3-1 (Z22.3) Camera Aperture. (1934 Chart 24) Tolerances 
have been added ; also the difference-dimensions E, F, and G, and the 
note below the table. 

DS35-4-1 (Z22.4) Projector Aperture. {1934 Chart 25) The dimen- 
sion between picture frames has been omitted. Tolerances have been 




7* A/r I 16-Mm. 16-Mm. 
35-Mm. | Double I single 

I 16-Mm. I 8M I Misc. 
I Single I 8 - Mm - I (m) 

Dimensional Standard* (DS) 


Cutting and perforating 







Projector Sprockets 





Camera Aperture 






Projector Aperture 






Emulsion and Sound- 

Track Positions in Cam- 







Emulsion and Sound- 

Track Positions in Pro- 







Sound Records and 

Scanned Areas 










Projection Lenses 



Unit of Photographic In- 




Lantern Slides 


Recommended Practice (RP) 







Release Print 



Screen Sizes 



Sound Transmission of 




Projection Room Layouts 






Photographic Density 



Projection Screen Bright- 







added, as well as the difference-dimensions D, E, F, G, and H. R has 
been increased slightly, and the note beneath the table added. 

Frame-Line. The center of the frame-line is midway between 
two successive perforations on each side of the film. 

DS35-5-1 (Z22.5) Emulsion and Sound Record Positions in Camera 
(Negative) . New Drawing. 

DS35-6-1 (Z22.6) Emulsion and Sound Record Positions in Projec- 
tor (Positive}. New drawing. 

DS35-7-1 (Z22.7) Sound Records and Scanned Area. (1934 Chart 
26) The width of the variable-width sound record has been increased 
from 0.71 to 0.76 inch, to correspond with present practice; also, 
tolerances have been added. 

DS35-8-1 (Z22.S) Reels. New drawing. 

DS35-9-1 (Z22.9) Projection Lenses. (a) No. 1 projection lens: 
External diameter of lens barrel 51.59 mm. (2Vs2 inches); (b) No. 2 
projection lens: External diameter of lens barrel 70.65 mm. (2 25 / 
inches). (No change from previous standard.) 

16-Mm. Doubly Perforated (DS16d) 

DS16d-l-l (Z22.10) Cutting and Perforating Negative and Positive 
Raw Stock. (1934 Chart 17) Tolerances for C and D have been in- 
creased to accord with European standards; also, G has been in- 
creased slightly. 

DS16d-2-l (Z22.ll) Sprockets. (1934 Charts, 20, 21, and 22) Re- 
vised drawing. No changes in data. 

DS16d-3-l (Z22.12) Camera Aperture. {1934 Chart 11) The differ- 
ence-dimensions C, E, and F have been added, as also the tolerances. 

DSl6d-4-l (Z22.13) Projector Aperture. (1934 Chart 11) Toler- 
ances have been added, as also the difference-dimensions C, D, E, and 

DSl6d-5-l (Z22.14) Emulsion Position in Camera (Negative). 
New drawing. 

DS16d-6-l (Z22.15) Emulsion Position in Projector (Positive). 
New drawing. 

DSl6d-8-l (Z22.16) Reels New drawing. 

16-Mm. Singly Perforated (DS16s) 

DS 16s- 1-1 (Z22.17) Cutting and Perforating Negative and Positive 
Raw Stock. New drawing, the dimensions given being the same as 


corresponding dimensions on DSl6d-l-l. (See remarks under 

ZSl6s-2-l Sprockets. In preparation. 

DS16s-3-l (Z22.18) Camera Aperture. (1934 Chart 28) Tolerances 
have been added, as also the difference-dimensions C, E, and F. 

DS16s-4-l (Z22.19) Projector Aperture. (1934 Chart 29) Tolerances 
have been added, as also the difference-dimensions C, D, E, and F. 

Frame-Line. The center of the frame-line shall pass through the 
center of a perforation on each side of the film. 

DS16s-5-l (Z22.20) Emulsion and Sound Record Positions in Cam- 
era (Negative). New drawing. The distance between the center of 
the picture and the corresponding sound has been increased from 25 
frames to 26 frames, in accordance with the action of the International 
Standards Association at the Budapest Conference in 1936. 

DS16s-6-l (Z22.21) Emulsion and Sound Record Positions in Pro- 
jector (Positive) . New drawing. The distance between the center of 
the picture and the corresponding sound has been increased from 25 
frames to 26 frames. 

DS16s-7-l (Z22.22) Sound Records and Scanned Area. (1934 
Chart 30) For the variable-width record, the width of the printed area 
has been decreased from 0.096 to 0.085 inch, and the width of the 
sound record has been increased from 0.060 to 0.064 inch. 

For the variable-density record the width of the sound record has 
been increased from 0.080 to 0.085 inch. 

The width of the scanned area has been increased from 0.065 to 
0.074 inch. Also, tolerances have been added. 

Film-track in Cameras and Projectors. A clearance of 0.13-mm. 
(0.005 inch) shall be allowed in designing the film-track in cameras 
and projectors. 

DS16s-8-l (Z22.16) Reels. Same as DSl6d-8-l (Z22.16). 

8-Mtn. (Ds8) 

DS8-1-1 (Z22.23) Cutting and Perforating Negative and Positive 
Raw Stock. New drawing. 

DS8-2-1 (Z22.24) Sprockets. New drawing. 

DS8-3-1 (Z22.25) Camera Aperture. New drawing. 

DS8-4-1 (Z22.26) Projector Aperture. New Drawing. 

DS8-5-1 (Z22.27} Emulsion Position in Camera (Negative). New 


(DS8-6-1) (Z22.28) Emulsion Position in Projector (Positive). 
New drawing. 
DS8-8-1 (Z22.29) Reels. New drawing. 


DSm-1-1 (Z22.30) Unit of Photographic Intensity. The. unit of 
photographic intensity adopted by the International Congress of 
Photography in 1933 shall be adopted for negative materials. (No 
change from previous standard.) 

DSm-2-1 (Z22.31) Lantern Slides. The mat opening shall be 3.0 
inches (76.20 mm.) wide by 2.35 inches (59.69 mm.) high. The 
thumb mark shall be located in the lower left-hand corner adjacent 
to the reader when the slide is held so that it can be read normally 
against the light. (No change from previous standard.) 

35-Mm. (RP35) 

RP35-1-1 (Z22.32) Film Splices, Negative and Positive.- (1934 
Chart 18) For the negative splices the dimension B in the 1934 edition 
was incorrect and is here corrected. Dimension F has been corrected 
accordingly. New dimensions C, D, and E have been added. For 
the regular positive and the full-hole positive splices, new dimensions 
C, D, and F have been added. 

RP35-2-1 (Z22.33) Release Prints. In course of revision. 

RP35-3-1 (Z22.34) Screen Sizes. Sizes of screens shall be in ac- 
cordance with Table II. 

The spacing of grommets shall be 6 inches, with 12 inches as a pos- 
sible sub-standard. The ratio of width to height of screens shall be 
4 to 3. (See /. Soc. Mot. Pict. Eng., June, 1933, p. 510.) 

The width of the screen should be equal to approximately y 6 th 
the distance from the screen to the rear seats of the auditorium. The 
distance between the front row of seats and the screen should be ap- 
proximately 0.87 foot for each foot of screen width. 

RP35-4-1 (Z22.35) Sound Transmission oj Screens. A loss of 2.5 
db. as given by the average response curve at 6000 cps. relative to the 
1000-cycle response as recorded, is a desirable limiting value for exist- 
ing types of sound equipment. Screens that meet this requirement 
are usually found to attenuate 4 db. at 10,000 cps. As to regularity 
of response, variations greater than 2 db. would not be tolerable. 



[J. S. M. p. E. 

No limits for regularity have been established for frequencies lower 
than 300 cps. (See /. Soc. Mot. Pict. Eng., Sept., 1931, p. 446.) 
RP35-5-1 (Z22.36) Projection Room Plans. Complete plans for 
projection rooms are contained in the Report of the Projection Prac- 
tice Committee published in /. Soc. Mot. Pict. Eng., Oct., 1935, 
p. 341. 

Screen Sizes 

Size No. 
of Screen 


Picture Height, 
Feet Inches 

Sire No. 
of Screen 































































































































Projection Lens Height. The standard height from the floor to 
the center of the projection lens of a motion picture projector should 
be 48 inches. 

Projection Angle. Should not exceed 12 degrees. 

Observation Port. Should be 12 inches wide and 14 inches high, 
and the distance from the floor to the bottom of the openings shall 
be 48 inches. The bottom of the opening should be splayed 15 de- 
grees downward. If the thickness of the projection room wall 
should exceed 12 inches, each side should be splayed 15 degrees. 

Projection Lens Mounting. The projection lens should be so 
mounted that the light from all parts of the aperture shall traverse 
an uninterrupted part of the entire surface of the lens. 

Projection Lens Focal Length. The focal length of motion picture 
projection lenses should increase in l / 4 -inch steps up to 8 inches, 
and in Vs-mch steps from 8 to 9 inches. 

Projection Objectives, Focal Markings. Projection objectives 
should have the equivalent focal length marked thereon in inches, 
quarters, and halves of an inch, or in decimals, with a plus (+) or 


minus ( ) tolerance not to exceed 1 per cent of the designated focal 
length also marked by proper sign following the figure. 

16-Mm. Doubly Perforated (RP16d) 

RP16d-l-l (Z22.37) Film Splices, Negative and Positive. (1934: 
Chart 12) Several difference-dimensions have been added in order 
to render the drawing clearer. 

16-Mm. Singly Perforated (RP16s) 

RP16s-l-l (Z22.38) Film Splices, Negative and Positive. New 


RPm-1-1 (Z22.39) Sensitometry. The principle of non-intennit- 
tency shall be adopted as recommended practice in making sensito- 
metric measurements. 

RPm-2-1 (Z22.40) Photographic Density. The integrating sphere 
shall be used as a primary instrument for the determination of photo- 
graphic density. Photographic densities determined by means of 
this primary instrument shall be used as secondary or reference stand- 
ards by means of which densitometers of other types may be cali- 

RPm-3-1 (Z22.41) Projection Screen Brightness. It is recommended 
that the brightness at the center of a screen for viewing motion pic- 
tures be between 7 and 14 foot-lamberts, when the projector is run- 
ning with no film in the gate. (See Bibliography: Report of Pro- 
jection Screen Brightness Committee.) 

RPm-4-1 (Z22.42) Nomenclature. A general glossary of technical 
terms used in the motion picture industry was published in /. Soc. 
Mot. Pict. Eng., Nov., 1931, p. 819; a glossary of color photography in 
May, 1935, p. 432; and a supplementary color glossary in Aug., 1936, 
p. 164. 


Number of Teeth in Mesh. The number of teeth in mesh with the 
film (commonly referred to as "teeth in contact") shall be the number 
of teeth in the arc of contact of the film with the drum of the sprocket 
when the pulling face of one tooth is at one end of the arc. 

Safety Film. The term "Safety Film," as applied to motion pic- 
ture materials, shall refer to materials having a burning time greater 
than 1,0 seconds and falling into the following classes: (a) support 


coated with emulsion, (b) any other material upon which or in which 
an image can be produced, (c) the processed products of these ma- 
terials, and (d) uncoated support that is or can be used for motion 
picture purposes in conjunction with the aforementioned classes of 

The burning time is defined as the time in seconds required for the 
complete combustion of a sample of the material 36 inches long, the 
determination being according to the procedure of the Underwriters 
Laboratory. This definition was designed specifically to define Safety 
Film in terms of the burning rate of the commercial product of any 
thickness or width used in practice. The test of burning time, there- 
fore, shall be made with a sample of the material in question having 
a thickness and width at which the particular material is used in 

All 16- and 8-mm. film must be of the safety type. 

E. K. CARVER, Chairman 












(All references are to J. Soc. Mot. Pict. Eng.) 

Reports of the Committee on Standards and Nomenclature: 
XV (Aug., 1930), No. 2, p. 160. 

Safety Code for Projection; Wide-Film Dimensions. 
XV (Dec., 1930), No. 6, p. 818. 

Projector and Camera Speeds; Standard Release Print; Screen 
Brightness; Negative Notching ; Wide-Film Dimensions. 
XVII (Sept., 1931), No. 3, p. 431. 

Wide-Film Dimensions. 
XVII (Nov., 1931), No. 5, p. 819. 

Glossary of Technical Terms Used in the Motion Picture Industry. 

XIX (Nov., 1932), No. 5, p. 477. 

16-Mm. Standards. 

XX (June, 1933), No. 6, p. 505. 
XXII (Jan., 1934), No. 1, p. 17. 


Standard SMPE Film Perforation; Unit of Photographic Intensity; 
Principle of Intermittency in Sensitometric Measurements. 

XXIII (Nov., 1934), No. 5, p. 247. 

Standards Adopted by the SMPE. 

XXIV (Jan., 1935), No. 1, p. 16. 

Stresa (Italy) Conference on 16-Mm. Standards. 

XXV (July, 1935), No. 1, p. 97. 

Report on Reichsfilmkammer at Berlin, April 25, 1935. 
XXV (Aug., 1935), No. 2, p. 192. 

XXV (Oct., 1935), No. 4, p. 370. 

International Standards Association Questionnaire Regarding 
16-Mm. Sound-Film Standards. 

XXVI (Jan., 1936), No. 1, p. 18. 
XXVI (May, 1936), No. 5, p. 597. 

Great Britain Adopts SMPE 16-Mm. Standards. 
XXVIII (Jan., 1937), No. 1, p. 21. 

XXVIII (June, 1937), No. 6, p. 585. 
16-Mm. Reduction Printing. 

XXIX (Oct., 1937), No. 4, p. 376. 

Report on Perforation Standards. 

Reports of the Projection Practice Committee (Selected) : 
XVII (Aug., 1931), No. 2, p. 245. 

Projection Room Lay-Outs. 

XXI (Aug., 1933), No. 2, p. 89. 

SMPE Standard Test Reels; Projector Optical Alignment. 

XXII (March, 1934), No. 3, p. 173. 

SMPE Standard Test Reels. 
XXII (June, 1934), No. 6, p. 379. 

Reel Lengths. 
XXV (Oct., 1935), No. 4, p. 341. 

Projection Room Lay-Outs (Revision). 
XXIX (July, 1937), No. 1, p. 39. 

Screen Brightness; Projector Apertures; Theater Survey; Screens. 
XXIX (Dec., 1937), No. 6, p. 614. 

Projector Apertures. 

Reports of Projection Screen Brightness Committee: 

XXV (Sept., 1935), No. 3, p. 269. 

XXVI (May, 1936), No. 5, p. 489-579. 

Symposium on Screen Brightness Problems. 

XXVII (Aug., 1936), No. 2, p. 1. 

Reports of the Sound Committee: 
XXVI (Jan., 1936), No. 1, p. 21. 

SMPE Primary and Secondary Frequency. 
Reference Standards. 

XXVIII (Jan., 1937), No. 1, p. 24. 


Reports of Non-Theatrical Equipment Committee: 
XXVH (Aug., 1936), No. 2, p. 161. 

16-Mtn. Screen Sizes and Illumination. 

XXVIII (Jan., 1937), No. 1, p. 26. 
16-Mm. Reels. 

XXIX (July, 1937), No. 1, p. 57. 

Reports of Color Committee: 

XXIV (May, 1935), No. 5, p. 422. 

A Glossary of Color Photography. 

XXVII (Aug., 1936), No. 2, p. 164. 

Supplementary Color Glossary. 
XXIX (July, 1937), No. 1, p. 54. 
Perforation Standards. 

Reports of Committee on Laboratory Practice: 
XXVI (April, 1936), No. 4, p. 345. 

Descriptive Report on Current Methods and Practices. 

Reports of the Projection Screens Committee: 
XVII (Sept., 1931), No. 3, p. 437. 

Acceptance Tests for Projection Screens. 
XX (June, 1933), No. 6, p. 510. 

Screen Sizes; Reflection Coefficients. 


XX (Feb., 1933), No. 2, p. 142. 

Lantern Slides and Scientific Charts. 

XXI (Oct., 1933), No. 4, p. 280. 

Historical Summary of SMPE Standardization. 

XXVIII (March, 1937), No. 3, p. 265. 
Camera Synchronizing Systems. 

Mar., 1938] 




(Sept. 20, 1930) 

For 35 mm Film 



Original: prior to 1928 
This revision: 193C 


J"^ - x " v 


- - 
















. r 






G --^3- 












Inch Equivalents 




35.00 -fO.OO 
- 0.05 
4.750 * 0.013 
2.794 * 0.01 
1.98 0.01 
3.40 * 0.05 
28.17 0.05 
Not > 0.025 
475.00 0.38 

1.378 + 0.000 
- 0.002 
0.1870 == 0.0005 
0.1100 0.0004 
0.0780 * 0.0004 
0.134 0.002 
1.109 0.002 
Not > 0.001 
18.700 0.015 

* L = the length of any 100 consecutive perforation intervals. 
** This single style of perforation, known as the SMPE perforation, shall 
be used for all 35 mm. film. It is the same as the perforation known prior to 
July 14, 1933, as the standard positive perforation. 

These dimensions and tolerances apply to the material immedi- 
ately after cutting and perforating. 



[J. S. M. P. E. 



(Sept. 20, 1930) 

For 35 mm Film 



Feed Sprocket 

Intermittent Sprocket 

Take-Up (Hold-back) Sprocket 











27 . 36 . 03 
24.00 0.03 
1.40 + 0.00 
- 0.05 
1.40 0.00 
- 0.05 

1.097 0.001 
0.945 0.001 
0.055 + 0.000 
- 0.002 
0.055 + 0.000 
- 0.002 

27.86 0.03 
24 . 00 . 005 
1 . 40 + . 00 
- 0.05 
1 . 40 + . 00 
- 0.05 

1.097 0.001 
0.945 0.0002 
0.055 +0.000 
- . 002 
0.055 + 0.000 
- 0.002 

27.86 0.03 
23.67 0.03 
1 . 40 + . 00 
- 0.05 
1 . 40 + . 00 
- 0.05 

1.097 0.001 
0.932 001 
0.055 +0 000 
- 0.002 
0.055 +0.000 

22 Deg. 30 Min. 1.5 Min. 

22 Deg. 30 Min. 0.75 

22 Deg. 30 Min. 1.5 Min. 

Recommended Practice 








0.077 1.96 
0.004 0.10 
0.935 23.42 
0.139 3.53 
0.040 1.02 
1.045 26.21 


* The accumulated error between any two teeth not to exceed 4 minutes. 

Mar., 1938] 




For 35 mm Sound Film 


Original; 1932 
This revision: 1936 


r ys_x 


^"^ ^ 








t_ J 



A , ^ 


























r a 





- i ^.^ 






Inch Equivalents 



22.05 0.05 
16.03 0.05 
18.90 0.05 
0.8 approx. 

0.868 =t 0.002 . 
0.631 =*= 0.002 
0.744 0.002 
0.03 approx. 

a = b = Vz longitudinal perforation pitch. 

The aperture dimensions given, in combination with the projector 
aperture shown in DS-35-4-1, result in a screen picture having a 
height- to- width ratio of 3 X 4 when the projection angle is 14 degrees. 

These dimensions and locations are shown relative to unshrunk 
raw stock. 



[J. S. M. p. E. 


For 35 mm Sound Film 

Original: 1931 
This revision: 1932 







" ^ 










i A * 






F IM/ 








[ j 







4. OF HLM 



Inch Equivalents 


20.95 0.05 
15.25 * 0.05 
18.74 0.05 
1.3 approx. 

0.825 0.002 
0.600 * 0.002 
0.738 == 0.002 
0.05 approx. 

a = b = l / t longitudinal perforation pitch. 

The aperture dimensions given result in a screen picture having 
a height-to-width ratio of 3 X 4 when the projection angle is 14 

These dimensions and locations are shown relative to unshrunk 
raw stock. 

Mar., 1938] 




For 35 mm Sound Film 

Original: 1930 
This revision: 1936 



Drawing shows film as seen from inside the camera looking 
toward the camera lens. 

(1) Emulsion position in camera: toward the lens, except for 

special processes. 

(2) Speed : 24 frames per second. 

(3) Distance between center of picture and corresponding sound: 

20 frames. 



[J. S. M. p. E. 



For 35 mm Sound Film 

Original: 1930 
This revision: 1936 

For Direct Front Projection 


Drawing shows film as seen from the light-source in the projector. 

(1) Emulsion position in projector: toward the light-source, except 

for special processes. 

(2) Speed: 24 frames per second. 

(3) Distance between center of picture and corresponding sound : 

20 frames. 

Mar., 1938J 





For 35 mm Sound Film 


Original: 1930 
This revision: 1936 







I P 





2.54 j_ 0.08 mm 


0.100" * 0.003 
1.93 0.25 mm 



0.076" == 0.001 
6.1 7 + 0.08 mm 



* s 1 


.-- - 

4 s< 







K 0.243" + 0.003 
2.54 * 0.08 mm 

^ > f 



0.100" =* 0.003 
6.1 7 + 0.08 mm 




t s 

^-0.243' + 0.003 



X ^ 

I a 

\ ^^^^^ 

2.1 3 0.025 mm 



0.084" * 0.001 
6.1 7 0.025 mm 

^"^^^^^^^^^^^ ^^ 


^0.243" 0.001 

When the push-pull type of sound record is used, the minimum 
separation between the two halves of the sound record shall be 0.152 
mm. (0.006 inch). When the squeeze-track is used with the variable- 
density record, the width of the sound record shall be 1.93 mm. 
(0.076 inch). 

These dimensions and locations are shown relative to unshrunk 
raw stock. 





For 35 mm Film 




300 Meters 

1000 Feet 

600 Meters 

2000 Feet 








8.3 Min. 

0.328 Min. 

Recommended Practice 






* This dimension applies only within a radius of 0.5 inch from the axis of the 

Mar., 1938] 



(Sept. 20, 1930) 

For 16 mm Silent Film 

Original: prior to 1928 
This revision: 1936 






Inch Equivalents 



16.00 -j-0.00 
- 0.05 
7.620 0.013 
1.83 0.01 
1.27 0.01 
1.83 0.05 
12.320 0.025 
Not > 0.025 
762.00 0.76 

0.630 +0.000 
- 0.002 
0.3000 0.0005 
0.0720 0.0004 
0.0500 * 0.0004 
0.072 0.002 
0.485 * 0.001 
Not > 0.001 
30.0 0.03 

* L = the length of any 100 consecutive perforation intervals. 

These dimensions and tolerances apply to the material immediately 
after cutting and perforating. 




(Aug. 28, 1935) 

For 16 mm Film 


DS 16d-2-l 

Number of Teeth in Mesh 





Combina-l Take-Up F . 
tion | (Holdback) 












600' 0.5' 
450' =*= 0.5' 
300' =t 0.5' 
2230' 0.5' 
























600' 0.5' 
450' 0.5' 
300' 0.5' 
2230' 0.5' 




. 036 








::s 4.", 


600' 0.5' 
450' 0.5' 
300' 0.5' 
2230' 0.5' 














Inch Equivalents 

N = Number of teeth on sprocket. 
Tolerance for B and C +0.000 to -0.025mm. 
or +0.000 to -0.001 in. 
Dimensional standards indicated by capital 
Recommended practice indicated by lower 
case letters. 
Values of C are omitted in cases where the 
angle of wrap on the sprocket would exceed 

12.22 + 0.05 
- 0.00 
1.22 + 0.00 
- 0.08 
B-0.3, Max. 
B + 1.52, 

0.481 + 0.002 
- 0.000 
0.048 + 0.000 
- 0.003 
. 050 
B-0.01, Max. 
B + 0.060, 

Mar., 1938] 



(Sept. 20, 1930) 

For 16 mm Silent Film 

Original: 1930 
This revision: 1936 











Inch Equivalents 



10.41 =b 0.05 
7.47 0.05 
8.00 =*= 0.05 
0.5 approx. 

0.410 0.002 

0.294 0.002 
0.315 =*= 0.002 
. 02 approx. 

a = b = Vz longitudinal perforation pitch. 

These dimensions and locations are shown relative to unshrunk 
raw stock. 



[J. S. M. P. E. 

(Sept. 20, 1930) 

For 16 mm Silent Film 


Original: 1930 
This revision: 1936 




Inch Equivalents 


9.65 * 0.05 
7.21 0.05 
8.00 0.05 
. 5 approx. 

0.380 * 0.002 
0.284 * 0.002 
0.315 * 0.002 
0.02 approx. 

a = b = x /2 longitudinal perforation pitch. 

These dimensions and locations are shown relative to unshrunk 
raw stock. 

Mar., 1938] 





For 16 mm Silent Film 

Original: prior to 1928 
This revision: 1936 




Drawing shows film as seen from inside the camera looking 
toward the camera lens. 

(1) Emulsion position in camera: toward the lens, except for 

special processes. 

(2) Normal speed: 16 frames per second. 



[J. S. M. P. E. 


For 16 mm Silent Film 

Original: prior to 1928 
This revision: 1936 

For Direct Front Projection 







Drawing shows film as seen from the light-source in the projector. 

(1} Emulsion position in projector: toward the lens, except for 

special processes. 
(2) Normal speed: 16 frames per second. 

Mar., 1938] 




For 16 mm Film 





120 Meters 

400 Feet 

240 Meters 

800 Feet 

48oflpf t 

1600 Feet 


Inch Equivalents 


Inch Equivalents 


Inch Equivalents 

8.10 + 0.00 
- 0.08 
8.10 + 0.00 
- 0.08 
17.2 Min. 

0.319 + 0.0008.10 + 0.00 
- 0.003 - 0.08 
0.319 + 0.0008.10 + 0.00 
- 0.003 - 0.08 
0.677 Min. 17.2 Min. 

0.319 + 0.000 
- 0.003 
0.319 + 0.000 
- 0.003 
0.677 Min. 

8.10 + 0.00 
- 0.08 
8.10 + 0.00 
- 0.08 
17.2 Min. 

0.319 + 0.000 
- 0.003 
0.319 + 0.000 
- 0.003 
0.677 Min. 

Recommended Practice 









NOTE: Center Spindle Holes Either a combination of square and round holes or two square 
holes may be used. 




(Aug. 28, 1935) 

For 16 mm Sound Film 

Original: 1932 
This revision: 1936 






Inch Equivalents 



16.00 + 0.00 
- 0.05 
7.620 0.013 
1.83 * 0.01 
1.27 0.01 
1.83 =*=0.05 
762.00 * 0.76 

0.630 +0.000 

- 0.002 
0.3000 * 0.0005 
0.0720 * 0.0004 
0.0500 * 0.0004 
0.072 * 0.002 
30.00 * 0.03 

*L = the length of any 100 consecutive perforation intervals. 

These dimensions and tolerances apply to the material immediately 
after cutting and perforating. 

Mar., 1938] 




(Aug. 28, 1935) 

For 16 mm Sound Film 


Original: 1932 
This revision: 1936 













Inch Equivalents 



10.41 =fc 0.05 
7.47 0.05 
8.00 =*= 0.05 
. 5 approx. 

0.410 =*= 0.002 
0.294 0.002 
0.315 * 0.002 
0.02 approx. 

a = b = l /t longitudinal perforation pitch. 

These dimensions and locations are shown relative to unshrunk 
raw stock. 



[J. S. M. P. E. 

(Aug. 28, 1935) 

For 16 mm Sound Film 


Original: 1932 
This revision: 1936 




Inch Equivalents 



9.65 0.05 
7.21 0.05 
8.00 0.05 
0.5 approx. 

0.380 0.002 
0.284 =fc 0.002 
0.315 0.002 
. 02 approx. 

a = b = 1 / 2 longitudinal perforation pitch. 

These dimensions and locations are shown relative to unshrunk 
raw stock. 

Mar., 1938] 




For 16 mm Sound Film 


Original: 1932 
This revision: 1936 


Drawing shows film as seen from inside the camera looking toward the 
camera lens. 

(1) Emulsion position in camera: toward the lens, except for 

special processes. 

(2) Speed : 24 frames per second. 

(3) Distance between center of picture and corresponding sound: 

26 frames. 



[J. S. M. P. E. 



For 16 mm Sound Film 



Original: 1932 
This revision: 1936 

For Direct Front Projection 



Drawing shows film as seen from the light-source in the projector. 

(1) Emulsion position in projector: toward the lens, except for 

special processes. 

(2) Speed: 24 frames per second. 

(3) Distance between center of picture and corresponding sound: 

26 frames. 

Mar., 1938] 




For 16 mm Sound Film 


Original: 1932 
This revision: 1936 


1.47*0.02 mm 





0.058" * 0.001 

I J 



2.15 * 0.05 mm 


0.085" * 0.002 
1.62*0.025 mm 



0.064" * 0.001 
1.47 * 0.02mm 







T --^ 



0.058" * 0.001 
2.15 * 0.05 mm 


0.085" * 0.002 


1 .47 * 0.02 mm 




I I \///////// 

-r i 

0.058" * 0.001 
1 .88 * 0.02 mm 

1 J 



0.074" * 0.001 
1.47 * 0.02mm 







0.058" * 0.001 




These dimensions and locations are shown relative to 
unshrunk raw stock. 



[J. S. M. p. E. 




For 8 mm Film 






16.00 +0.00 

0.630 + 0.000 

- 0.05 

- 0.002 


3.810 0.013 

0.150 0.0005 


1.83 0.01 

0.072 0.0004 


1.27 0.01 

0.0500 0.0004 


1.83 0.05 

0.072 0.002 


12.320 0.025 

0.485 0.001 


Not > 0.025 

Not > 0.001 


8.00 +0.00 

0.315 + 0.000 

- 0.08 

- 0.003 


381.00 0.38 

15.0CO 0.015 




* L = the length of any 100 consecutive perforation intervals. 

These dimensions and tolerances apply to the material immediately after cut- 
ting and perforating. 

Film may be slit before or after processing. 

Mar., 1938] 




For 8 mm Film 










5.72 == 0.03 
9.42 + 0.00 
- 0.05 
1.02 + 0.00 
- 0.05 
1.14 + 0.08 
- 0.00 

450' = 

0.225 0.001 
0.371 + 0.000 
- 0.002 
0.040 + 0.000 
- 0.002 
0.045 + 0.003 
- 0.000 

t 0.5' 

Recommended Practice 






S. M. P. E. 



For 8 mm Silent Film 




<fe OF FILM 



Inch Equivalents 


4.80 =*= 0.03 
3.51 0.03 
5.22 0.05 

0.189 * 0.001 
0.138 * 0.001 
0.205 0.002 

a = b = Va longitudinal perforation pitch. 

Mar., 1938] 




For 8 mm Silent Film 









4.37 0.03 
3.28 0.03 
5.22 0.05 

0.172 =*= 0.001 
0.129 * 0.001 
0.2055 * 0.002 

a = b = Va longitudinal perforation pitch. 




For 8 mm Silent Film 




Drawing shows film from inside the camera, looking toward the camera lens. 

(1) Emulsion position in camera : toward the lens, except for special 

(2) Normal speed: 16 frames per second. 

Mar., 1938] 




For 8 mm Silent Film 



For Direct Front Projection 



Drawing shows film as seen from the light-source in tte projector. 

(1) Emulsion position in projector: toward the lens, except for 
special processes. 

(2) Normal speed: 16 frames per second. 



[J. S. M. P. E. 



For 8 mm Silent Film 




Capacity, 60 M. (200 Ft.) 


Inch Equivalents 


8.10 + 0.00 
- 0.08 

0.319 + 0.000 

- 0.003 
0.35 Min. 

Recommended Practice 






Drive side of sprocket may have any desired odd number of driv- 
ing slots, evenly spaced. 

Mar., 1938] 




For 35 mm Sound Film 

Original: 1928 
This revision: 1936 












Regular Positive 

Full Hole Positive 


Inch Equiv. 


Inch Equiv. 


Inch Equiv. 












[J. S. M. P. E. 

(Sept. 20, 1930) 

For 16 mm Silent Film 


Original: 1930 
This revision: 1936 









Inch Equiv. 


Inch Equiv. 






Mar., 1938] 




(Aug. 28, 1935) 

For 16 mm Sound Film 


Original: 1932 
This revision: 1936 






Inch Equiv. 


Inch Equiv. 







Summary. Statement of the activities of the Standards Committee during the past 
six months, leading to completion of the revision of the SMPE standards published 
elsewhere in this issue of the Journal. 

There have been but two meetings of the Standards Committee 
since the last report was written. One of these was devoted to a 
final correction of a series of drawings that had been practically 
approved, and the second one dealt with the initial drafts of four- 
teen new drawings completed during the summer. These drawings 
covered 8-mm. film standards, a revision of the drawings for the 
35-mm. and 16-mm. projection sprockets, and reels for 35-mm., 
16-mm., and 8-mm. film. 

35-Mm. and 16-Mm. Projection Sprockets. In connection with the 
sprocket drawings, no essential changes in either design or dimensions 
have been suggested. 

Projection Reels. For the 35-mm. reels, both 1000- and 2000-ft. 
capacity, dimensions are being prepared. The 2000-ft. reel dimen- 
sions will agree essentially with the specifications proposed by the 
Academy, although only the basic dimensions will be given in the 
SMPE drawings. 

In the case of the 16-mm. reels, objection has been raised toward 
standardizing the reel with the square hole on one side and the round 
hole on the other side, so that the present proposal by the Com- 
mittee is to use two standards: the square-round combination to 
be used by those who wish to use it, and the square-square combi- 
nation to be used by those who prefer that. The main objection to 
the square-round combination is that this feature is patented. At 
its last meeting, the Committee did not feel justified in standardiz- 
ing exclusively on any patented feature of this sort. 

For the 8-mm. reels, standards are being prepared only for the 
projection reels. The camera reels vary according as the manu- 
facturer desires to use the double- width 8-mm. film in the camera 
and to slit after processing, or to use the single- width 8-mm. film 
the camera. 

* Presented at the Fall, 1937, Meeting at New York, N. Y 


Sound Records and Scanned Area, 35-Mm. Film. A ballot was 
taken during the summer on the following proposal : 

"In the case of the push-pull track, the space separating the two halves of the 
track shall be 0.15 mm. (0.006 inch) wide, and centered upon the center-line of 
the sound record." 

Although the balloting was in favor of this proposal, there were 
enough objections to the 0.006-inch width to warrant going further 
into the matter, especially as there is some disagreement as to the 
actual practice used in the trade. Mr. J. O. Baker has been ap- 
pointed a committee of one to investigate this problem thoroughly 
and to report back to the Committee. Anyone having definite 
ideas on the subject should communicate with the General Office of 
the society. 

Perforation Dimensions. The Committee has not forgotten the 
recommendation of its Sub-Committee on Perforation Size that the 
dimensions of the positive perforation be changed so as to coincide 
more closely with those of the old Bell & Ho well perforation. Work 
was started last spring on a punch and die to be built in accordance 
with the specifications of Howell & Dubray. Owing to press of 
other work, and to the desire to have these dimensions as accurate 
as would be obtained in commercial work, special cams were de- 
signed and built, and the punches have been finished only within a 
few days. A thorough, practical test on film perforated with these 
dimensions is being undertaken by the Committee. 

Standardization of Densitometers. Owing to Dr. O. Sandvik's 
absence in Europe, no practical steps have been taken during the 
summer to prepare samples of film standardized by means of the 
integrating sphere for use in standardizing densitometers in the 
trade. The matter, however, will receive attention as soon as 

E. K. CARVER, Chairman 












Summary. A brief discussion of the new trends in studio set lighting increased 
use of dolly shots; use of lamps above the set instead of on the floor; improvements 
in lens and reflector types of lamps; greater intensities required for color cinematog- 
raphy; lighting sets according to "key lights," with much less "general" lighting. 
The report concludes with a description of some of the new set lighting equipment. 

If we look back over the history of almost any art or science, 
whether it be sound recording and reproduction or motion picture 
photography or any similar activity, we can not help observing an 
interesting correlation between the advances in the art and the de- 
velopment of new devices or tools. It is frequently difficult to decide 
whether the development of the new tools is the cause or the result of 
the progress being made. 

Advances in motion picture studio lighting during the past year or 
so provide an excellent example of this phenomenon. Here we have 
new types of lighting units, improvements in illuminants, the special 
requirements of lighting for color, new photographic emulsions, and 
the greater use of the moving-camera or "dolly" shot, all exerting 
their influence to bring on virtually a new era in the art of motion 
picture photography. 

No doubt the dolly shot, which is a very definite contribution to 
the continuity and smoothness of action of the motion picture story, 
is one of the earlier influences. This form of camera operation makes 
the use of a multiplicity of floor-lighting units out of the question, and 
has put the lamps almost entirely upon the lamp-rails above the set. 
Because of the greater distances over which the light must be di- 
rected when the equipment is mounted overhead, the so-called ' 'gen- 
eral lighting units, ' ' with their broad beam spreads and limited pene- 
trating power, have almost entirely given way to spotlighting equip- 
ment with accurately controllable beam spreads. 

With the greater emphasis toward "spots," their well-known 
shortcomings, such as low efficiency, spill light, non-uniform illumina- 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received October 
8, 1337. 


tion pattern, become intolerable. New spotlighting equipment, im- 
proved in these respects in both lens and reflector types, has al- 
ready been discussed 1 - 2 or will be mentioned later in this report under 
the heading of new equipment. 

The necessity of illumination intensities of 800 to 1000 foot-candles 
for the earlier Technicolor film emulsions caused the development 
of units of much higher power and efficiency, as, for example, the 
types 90 and 170 high-intensity arcs employing Fresnel lenses, so as 
not to increase unduly the number of units required. Paralleling the 
introduction of the big arcs, higher-powered and more efficient in- 
candescent lamp units, giving uniform illumination patterns, have 
been made available. These employ improved 2000-, 5000-, and 
10,000-watt lamps. Recently Fresnel lens spots of 500 and 1000 
watts have been developed. 

Thus the cameraman has the choice of a greater range of more 
adaptable lighting devices than ever before. At the same time, a 
smaller number of sources is required in lighting a given set, which 
has resulted in the use of strong highlights to accentuate the points 
of interest in the scene, sometimes called "key" lights. Often the 
remainder of the set is in comparative shadow. 

The effect upon the art of lighting has been pronounced. Lighting 
has in many instances become one of the features of the picture, some- 
times even obtrusive. Much of the excessive stylizing will doubtless 
prove to be a passing fad. Out of all the experimentation, however, 
new values are being introduced. 

Tony Gaudio, one of the old-time cameramen, yet among the 
most progressive, has introduced a new technic, which he chooses to 
call "precision lighting." With the aid of the light-control features 
of the new spots he lights only the chief points of interest, leaving 
the remainder of the set at a relatively low level of illumination. As 
the actor moves about, other spots previously adjusted to cover the 
actor's movements are brought to full brilliancy by means of dim- 
mers, and the lamps not then needed are gradually dimmed. This 
has the desirable effect of carrying the audience's attention with the 
actor at all times. It does, of course, call for unusual skill in direc- 
tion, a high order of coordination among director, cameraman, and 
electrical staff, and ample rehearsal ; but the result is an appealing 
and effective type of photography that will, no doubt, have its in- 
fluence on all cameramen. 3 

Besides serving to focus the attention of the audience upon desired 


parts of the picture, the cameraman, with the aid of his lighting, en- 
deavors to overcome the limitations of the two-dimensional picture in 
illustrating a three-dimensional scene. By backlighting he brings 
the actors into relief. In close-ups, modelling is achieved by several 
sources and by careful gradation of light-intensities across the faces. 
There are limitations to such technics beyond which the effect is an 
unnatural one. 

With the advent of photography in color the idea prevailed that 
the lighting could be much flatter, since the color would provide the 
element of depth now supplied by the lighting. Furthermore, there 
seemed to be no alternative, because of the limited latitude of ex- 
posure of the photographic emulsions used in color work. The range 
of illumination intensity between highlights and shadows possible in 
black-and-white photography would result in over- or underexposure 
when applied to a color material. That was true with the earlier 
Technicolor emulsions, but recent improvements in both the ma- 
terial and its subsequent processing have greatly extended the light- 
ing-contrast range of this particular color system. As a result, 
Technicolor cameramen are lighting with fully the same contrasts 
employed for black-and-white photography, with improved quality 
in the finished picture. 


In discussing new studio lighting equipment the Committee is not 
in any sense duplicating the work of the Progress Committee, 4 but 
feels that it can render the industry a service in appraising the value 
of the new devices and discussing their effect upon studio lighting 
practice. It is interesting to observe that the new equipment to be 
discussed really represents refinements or extensions of equipment 
already in use in the direction of more accurate light control. 

One manufacturer of lighting equipment during the past year has 
introduced a "Triple Five" studio spot. Recognizing the rather 
non-uniform field of illumination produced by parabolic-mirror re- 
flector spots, particularly at the wider beam spreads, this firm has 
placed a specially designed Fresnel condensing lens at the front of 
the lamp to confine light that otherwise would become objectionable 
spill light within the angles of the beam from the parabolic mirror. 
The effect is a marked increase in efficiency of utilization and an im- 
provement in quality, with the darker area in the center of the out- 
of-focus beam filled in. Since the lens adjustment necessary for a 


particular beam spread differs from that required with the parabolic 
mirror, the new lamp incorporates a differential leverage adjustment 
of such design that, as the lamp is moved into the mirror to gain a 
wider beam spread, the lens is moved toward the lamp at a somewhat 
different rate. Thus the beam of light coming from the lens at all 
times fills in the center of the spot produced by the parabolic mirror. 
This arrangement serves the double purpose of gaining a more 
uniformly illuminated field. The same firm also manufactures a 
compact 2-kw. lamp spot, employing the Fresnel type of condensing 
lens. In order to eliminate spill light from the risers of the lenses 
the risers have been blackened. 

Another lighting equipment manufacturer, also of Hollywood, has 
added three units to his already rather complete line of incandescent 
and arc lighting equipment. One is the type 65 high-intensity arc 
spot. This lamp employs a 9-mm. high-intensity positive carbon 
and a 5 /ie-inch diameter negative carrying a current of 65 amperes. 
The unit is considerably more compact than the type 120 and the 
150-ampere arc spots. It employs a smaller Fresnel-type lens and is 
intended to be used in the more restricted locations where larger 
spots can not be employed. The same firm has also made available 
two smaller incandescent lamp spots, one known as the type 206 
Solar Spot, employing the 500-watt G-30 bulb monoplane filament 
lamp with the new medium-bipost base; the other, type 208, uses 
the 1000- watt G-40 bulb medium-bipost base lamp. These spots 
are characterized by extreme compactness. They are intended pri- 
marily for close-ups in restricted localities where space is at a pre- 
mium. These units are frequently mounted directly upon the camera 
blimp for use in connection with dolly or travelling shots. Lamps 
for the units are available in both the black-and-white and color- 
photography ratings. 

The lamp manufacturers have introduced improvements in the ef- 
ficiency of the 2000-watt G-48 bulb lamp, the 5000-watt G-64 bulb 
and the 10, 000- watt G-96 bulb lamps as employed for black-and-white 
photography. This is in answer to a demand from many cameramen 
for a whiter light and maximum output from these sources. Their 
observations in connection with color photography have suggested 
the advantages of lamps of higher efficiency for all types of pro- 

In addition to these changes, the lamp manufacturers have in- 
troduced a new group of lamps paralleling in most instances the black- 


and- white types and known as the CP line. They have been de- 
signed for a uniform color temperature of 3380 K, and include the 
familiar 2000-watt Movieflood in the PS-52 bulb for use in general 
lighting equipment, a 2000-watt in the G-48 bulb, a 5000-watt in the 
G-64 bulb, and a 10,000-watt in the G-96 bulb. The last three are 
fitted with the mogul bipost base. The unusual feature of these 
lamps, distinguishing them from any other group of lamps designed 
for a particular service, is that the color of the light is the same for all 
wattages. This is in deference to the very close color requirements 
of the Technicolor process. Thus, when used with the proper filter 
for color photography, the color is the same throughout the entire 
set, and, in addition, can be mixed with properly filtered arc light 
or with daylight. The effect of the improved lighting units upon 
the number of units employed for set lighting has been well discussed 
by Handley and Richardson. 5 ' 6 

New Incandescent Lamp Filters. At the Fall meeting of the So- 
ciety in New York in 1934, the present chairman of the Studio Light- 
ing Committee presented a paper 7 describing two relatively simple 
filters that permitted satisfactory Technicolor photography with in- 
candescent lamps. These filters were not sufficiently accurate, how- 
ever, to allow mixing or interchanging indiscriminately several dif- 
ferent illuminants such as daylight, arcs, and incandescent lamps. 

During the past year more precise filters have been produced by 
employing a medium-blue glass base and spraying it with an enamel 
made up principally of cobalt alumina. In the firing process, the 
cobalt alumina, which is blue, partially changes to cobalt silicate, 
which is purple ; and by exact control of this feature the transmission 
characteristics of the filter are kept within very precise limits. The 
firing process also "tempers" or renders the glass non-shatterable.* 
These filters are now being used in regular Technicolor productions. 
Thus the Technicolor cameraman is provided with a range of lighting 
equipment fully as extensive as that available to the black-and-white 

Development in Other Sources. A survey of developmental work now 
in progress on the gaseous conductor lamps shows no new types that 
have not already been covered in previous reports of this Committee 
and of the Progress Committee. The high-intensity, water-cooled 
capillary mercury arc still appears to have the best possibilities of 

* Libbey-Owens-Ford Vitrolux. 


this type of illuminant for motion picture work. At present lamps 
of about 1000- watt rating are being manufactured experimentally, 
having efficiencies of the order of 60 or more lumens per watt and a 
source brightness of the order of 250,000 candles per square-inch and 
higher. Before such sources can be used for motion picture work, 
however, there still remain the problems connected with liquid cool- 
ing, and cyclic variation of the light with the current frequency, since 
high voltages are involved. The light quality of the present lamp 
is also not entirely suitable, particularly for color work, but improve- 
ments are being made in this direction. 

In the allied branches of studio lighting, such as power production, 
wiring methods, etc., there appears to have been little change in the 
past year. Mole-Richardson, Inc., has made available three up-to- 
date portable, gasoline -driven power plants rated at about 1400 
amperes each, employing d-c. generators developed especially to 
meet the load-speed characteristics of the gasoline engine and incor- 
porating high-speed voltage-control devices to prevent overshooting 
the voltage should part of the load be switched off. The housing 
surrounding the engine and the exhaust have been carefully designed 
to render them so nearly noiseless that they may be stationed within 
200 feet of the microphone. 

R. E. FARNHAM, Chairman 




1 RICHARDSON, E. C.: "A Wide-Range Spotlamp for Use with 2000-Watt 
Lamps," /. Soc. Mot. Pict. Eng., XXVI (Jan., 1936), No. 1, p. 95. 

2 RICHARDSON, E. C.: "Recent Developments in High-Intensity Arc Spot- 
lamps for Motion Picture Production," /. Soc. Mot. Pict. Eng., XXVIII (Feb., 
1937), No. 2, p. 206. 

3 GAUDIO, G.: "A New Viewpoint on the Lighting of Motion Pictures," /. 
Soc. Mot. Pict. Eng., XXIX (Aug., 1937), No. 2, p. 157. 

4 Report of the Progress Committee, /. Soc. Mot. Pict. Eng., XXIX (July, 
1937), No. 1, p. 3. 

5 HANDLEY, C. W.: "The Advanced Technic of Technicolor Lighting," /. 
Soc. Mot. Pict. Eng., XXIX (Aug., 1937), No. 2, p. 169. 

6 RICHARDSON, E. C.: "Recent Developments in Motion Picture Lighting," 
/. Soc. Mot. Pict. Eng., XXIX (Aug., 1937), No. 2, p. 178. 

7 FARNHAM, R. E.: "Recent Developments in the Use of Mazda Lamps for 
Color Motion Picture Photography," /. Soc. Mot. Pict. Eng., XXIV (June, 1935), 
No. 6, p. 487. 



Summary. A statement of recent activities of the Committee, and a proposed pro- 
gram of future work. 

The Committee met at the Hotel Pennsylvania, New York, N. Y., 
on October 13, 1937. Members in attendance were J. G. Bradley, 
Chairman, J. I. Crabtree, A. S. Dickinson, T. Ramsaye, and J. E. 
Abbott. Substitutes attending were K. Famulener for W. A. Schmidt 
and C. A. Lindstrom for R. Evans. Mr. Bradley substituted for 
C. L. Gregory. V. B. Sease and M. E. Gillette were represented by 
written submissions setting forth their views on problems before the 

The Chairman made a brief report on related work being done at 
Washington, summarized briefly as follows: (J) The National Ar- 
chives storage cabinets, previously approved by the Committee, have 
been installed, and field tests with other types of cabinet are being 
made; (2) the research work at the Bureau of Standards on pres- 
ervation of records has been resumed; and (3) the Federal Fire 
Council has become increasingly active in an effort to minimize fire 

It was pointed out that emphasis is shifting (temporarily, at least) 
from a consideration of preservation in terms of deterioration to a 
consideration of preservation in terms of film fires and film handling. 
This shift does not represent lack of interest in the problems of de- 
terioration, nor does it indicate that the work in that field has been 
completed. It is based rather upon the fact that a very large quan- 
tity of nitrocellulose material in Federal custody has been discovered 
recently, lacking adequate fire protection. It is also the result of 
increasing interest on the part of corporate and private collectors in 
preserving this type of record. Adding momentum to this shift 
have been some recent and rather extensive film fires. It was agreed 
that the Committee should take cognizance of the fire problem as 

*Presented at the Fall, 1937, Meeting at New York, N. Y.; received October 
27, 1937. 



falling in the field of preservation, and render all possible service 
to those seeking help. 

The need for additional funds with which to carry on this work was 
mentioned, and various means for meeting the emergency were dis- 
cussed. It was the consensus of opinion that, while certain private 
individuals and non-Federal agencies might be expected to contribute 
nominally, the U. S. Government should finance this work, being 
the most representative agency of all the people. 

The Chairman submitted a proposed work program to cover 
future activities of the Committee, which had been the subject of 
correspondence between him and the individual members. This 
program follows : 

(jf) The effects of projection heat and light upon films. 

(2) Continued investigation of the use of acetate film base for long-time stor- 

(5) Investigation of other film base materials not in current use. 

(4) A safe yet economical plan for long-time storage. 

(5) A study of the handling and winding of film roll winding under dust-free 

(6) Standardized physical and chemical tests to determine condition of films 
received for storage. 

(7) Chemical and physical treatment for restoration of faded images, deterio- 
rated emulsion, and decomposing base. 

(8} Specifications for processing film for long-time storage. 
(5) A study of printers for old shrunken film. 
(10} Definition of terms: a glossary. 

The proposed program was discussed at length and was finally 
adopted as a general guide, with the recommendation that it be 
made sufficiently flexible to meet current and local situations. The 
discussion also enlarged the scope of some of the foregoing items 
and limited the scope of others. For example, it was suggested that 
the Committee should continue its study of nitrocellulose film for 
the reason that for a long time to come it will be necessary to handle 
that type of material. As to a study of new materials for film bases, 
the view was expressed that the practical aspects of this properly 
belonged in the field of commercial enterprise rather than in the 
experimental field as found in the laboratory ; and that much experi- 
mentation lies ahead of any successful effort to gain wide acceptance 
of a new base all modified by market conditions, sales resistance, 
cost of materials, etc. The best service the laboratory can render 
in this regard is to examine specimens submitted to it and give out its 


unbiased findings. A study of emulsions, projection light and heat, 
storage and preservation, handling, restoration of faded images 
(when practicable), chemical tests, printers for old film, and a stand- 
ardized glossary were emphasized. 

The question was asked whether the Committee should act as a 
review board or whether it (or its individual members) should par- 
ticipate actively in certain phases of research. Mr. Crab tree ex- 
pressed the view, shared generally by other members and substi- 
tutes present, that while some of the members represented insti- 
tutions having facilities for doing research work, others did not; 
and, furthermore, that the final stamp of approval by the National 
Bureau of Standards would give general acceptance to the work not 
given to commercial institutions or private individuals. Hence 
it was agreed that the Committee should continue to act largely in an 
advisory capacity to the Bureau, except for the following division of 
labor among members of the Committee : 

(1) Mr. Crabtree agreed, with the help of Messrs. Sease and 
Schmidt, to assume the leadership in the matter of handling film as 
set out in item 5 of the work program. 

(2) Mr. Bradley agreed, with the help of Mr. Dickinson, to take 
over the work of storage and fire control as set out in item 4. 

(3) Item 9, a study of printers, was assigned to Messrs. Gregory 
and Abbott. 

(4) The matter of a standardized glossary as outlined in item 10 
was assigned to Messrs. Gregory and Ramsaye. 

It was agreed that all other items in the program not covered by 
these special assignments were to be left to the Chairman, with the 
help of Messrs. Gillette and Evans, with recommendations that the 
final work be cleared through the National Bureau of Standards. The 
Committee pledged its support to this work, and the members pres- 
ent representing manufacturers with research facilities extended 
this pledge to include the use of these facilities by the Bureau's tech- 
nical workers. 

J. G. BRADLEY, Chairman 





Summary. The volume of film being used for permanent record purposes is 
rapidly increasing. This includes principally the documentary type of motion 
pictures. However, better entertainment pictures are being produced which can reason- 
ably be expected to live just as a good book lives. Increased volume of documentary 
records and increased interest in permanency demands planned storage, both in te-rms 
of preservation and in terms of fire hazard. The principle of unit isolation is the 
basis for storage at The National Archives. Spread of film fire can be prevented by 
(1} insulation and (2) use of a cooling agent. Among cooling agents tested, water was 
found effective and economical. Cascade type of shelving holds out great promise. 

Three changing aspects of the film -storage problem seem pertinent 
to this discussion. Two of these are offered as evidence of the prob- 
lem, and one (the third) as a possible solution. The first is the in- 
creased use of photographic material, and the second is the increased 
interest in permanency. 


Not only are present users of film increasing their output but the 
number of users is increasing rapidly. Reference is made here to 
institutions making photographic records of the documentary type. 
For example, recent surveys have located nearly 1500 depositories of 
photographic records in federal custody alone. Some of these are 
large and some are quite small, but 282 of them have either 5000 or 
more still negatives or 10,000 feet or more of motion pictures. The 
total volume will perhaps exceed 600 tons of material. 

Various governmental agencies are engaged in making aerial maps, 
and hope some day to photograph every square mile of the United 
States. The Bureau of the Census is engaged in microfilming its 
enormous census records. The National Archives is giving serious 

*Presented at the Spring, 1937, Meeting at Hollywood, Calif.; received May 
13, 1937, revised November 4, 1937. 

** Chief, Division of Motion Pictures and Sound Recordings, The National 
Archives, Washington, D. C. 


304 J. G. BRADLEY [j. s. M. P. E. 

thought to the same technic as a method of ultimately reducing the 
millions of records in its custody. When the committee in charge of 
the film preservation work, 1 done at the National Bureau of Stand- 
ards, announced that this type of record could be expected to last a 
long time, interest in the production of such records by libraries, 
schools, and commercial firms was greatly stimulated. These are a 
few of the many examples that could be cited if further evidence 
seemed necessary, but it is reasonable to expect that the coming years 
will develop both new uses and new users of this very effective type of 


It is evident that some of the increased use of photographic material 
is based upon an interest in permanency, but this is not the sole fac- 
tor. The production of entertainment pictures, educational or in- 
structional pictures, or of documentary pictures for other purposes, 
may be entirely divorced from any thought of permanency. On the 
other hand, the interest in permanency seems to have its own founda- 
tion, and to be the result of a natural trend. 

Individuals and industries are a great deal alike in many respects. 
Youth offers one interest and mature years offer another. A young 
man taking his first job is interested (or should be) in making good at 
his work. If married, he builds a house as a shelter, buys clothes to 
cover his back, and eats simple food. In brief, he is a utilitarian. A 
little later he is interested in gaining recognition and making money. 
When his money is made and he has leisure, he takes up golf, buys 
etchings, and writes his memoirs. When a young industry starts 
out it is interested in hard work and in making good. Neither its 
policy nor its product has received general recognition. Perhaps both 
its capital and credit are limited. The basement or the old ware- 
house will do for quarters. It is interested in making the product and 
selling it, in beating last year's quota, in making money and lots of it. 
In brief, it is highly utilitarian and this is as it should be. Later, 
when it has made its money, cultural interests may find a place in its 

The trend on the part of the motion picture industry in this direc- 
tion seems well defined. Both the number and percentage of motion 
pictures dealing with artistic and cultural subjects are increasing. 
Classical literature is being drawn upon heavily as source material. 
The services of scholars are being enlisted to authenticate data. Sci- 


enlists, musicians, and eminent authorities in various fields are being 
called to Hollywood just as the engineer was called when sound was 
introduced. That this trend may rest upon profit motives is not ex- 
actly pertinent to this discussion. In any event, the time has come 
when we may expect a good picture to live just as we expect a good 
book to live ; for which we should be grateful. 

Aside from the hope that good entertainment motion pictures as 
produced by the industry may be preserved, there remains a very 
genuine and legitimate interest in preserving photographic records 
for legal, historical, educational, and governmental reasons. Land 
and other property claims continue sometimes over several genera- 
tions. Veterans' claims and claims under the new social security 
legislation will continue for a long time to come. For example, Har- 
vard University wishes to preserve the motion pictures it took at its 
tercentenary celebration so that they can be shown a hundred years 
hence. The Wanamaker heirs are making definite plans to perpetu- 
ate the valuable collection of American Indian motion pictures as a 
memorial to Rodman Wanamaker. People from all over the world 
from Australia to Russia have made personal calls at The National 
Archives, and others have written, asking how they can keep their 
film. And finally there is the great body of newsreels which few will 
dispute should be preserved. 

If this increased volume represents a fixed trend, and if the interest 
in preservation is genuine and permanent, then greater emphasis 
must be given to planned storage. No longer will the cellar or the 
abandoned garage be good enough. One is reminded here of auto- 
mobile parking; when there were few cars and traffic was light, any 
place along the curb was good enough; there was no problem. Now 
auto parking must be reckoned with, and photographic film, par- 
ticularly nitrocellulose film, demands basic planning. 


Although our research at The National Archives on storage in 
terms of preservation, to which members of the Society of Motion Pic- 
ture Engineers and others have made valuable contributions, is far 
from complete, we have learned a few things, and collected and coa- 
lesced other things already known, most of which had been presented 
in previous papers. Recently we have been giving considerable at- 
tention to storage in terms of fire-control. That work is also unfinished 
but certain preliminary results may prove interesting or even helpful. 



fj. S. M. p. E. 

In our plans for film-fire control unit isolation is emphasized above 
everything else. This principle, of course, is not new; we merely 
have adapted it to certain variations. But let us review the matter 

FIG. 1. The National Archives film -storage cabinet set up for 
fire test. Figures above the arrows indicate temperatures (Fahr.) 
obtained without the container; figures below the arrows, with 
the container. 

for a moment. A vault of 750 cubic-feet, holding approximately 
5000 pounds of nitrocellulose film on open racks, is generally accepted 
as the maximum. An intermediate step is an enclosed cabinet in 
which approximately 100 pounds (more or less) is the average unit of 

Mar., 1938] 



isolation. Finally, there is the compartment or pigeon-hole, with 5 
pounds as the unit of isolation. This latter form of storage was re- 
ported by Crab tree and Ives 2 some time ago. The National Ar- 
chives has adopted this principle for its more valuable motion pic- 
tures, but has gone considerably further in the matter of insulation. 


Lower Compartment 

10 20 30 40 50 60 
Upper Compartment 

10 20 30 40 

50 60 

FIG. 2. Time-temperature curves of fire 
test of motion picture film cabinet. 

Lower compartment: Curve 1, on top of can. 
Curve 2, inside can on top of film. 

Upper compartment: Curve 1, under can. 
Curve 2, inside can on top of film. 

Although the design of this cabinet has been reported, 1 the fire test 
was not made until recently. 

Fire Test on Insulated Cabinet. (See Fig. 1.) The test was made 
with one unit of a sample cabinet furnished by the General Fireproof - 
ing Company of Youngstown, Ohio, under their contract with the 
Procurement Division of the Treasury Department for The National 
Archives. The unit consisted of ten horizontal-drawer compart- 

308 J. G. BRADLEY tf. a M. p. E. 

ments and a vertical flue, a base section, and a top section. The 
cabinet was fabricated from corrosion resistant steel throughout, and 
all walls, drawer-fronts, and spaces between compartments were in- 
sulated. The insulation was approximately 1 inch thick, leaving a 
12 l /2 X 12 3 /ie-inch compartment (inside measurements) 2*/2 inches 
high, and a 12 Y 2 X 4-inch flue in the rear. The drawer heads were 
of the icebox type and were held in place under pressure. A gravity 
flap with breather ports 1 separated the compartment from the flue. 

In the preparation of the fire test, the base section was laid in a bed 
of plaster on a flat piece of plaster board. Rebates between the sec- 
tions were filled with putty (Alumilastic, consistency "C") before fit- 
ting the sections together. 

Three full reels of old nitrate film were placed in the special vented 
containers developed by The National Archives 3 and located in com- 
partments 6, 7, and 8, reading from the bottom drawer as number 1. 
An ignition coil, connected to a source of current, was placed around 
the outer edge of the reel in compartment 7. Thermocouples for 
measuring temperatures were located in all three drawers. 

The outside temperature was approximately 51F. Fumes were 
noted issuing from the top of the cabinet 10 seconds after electric con- 
tact was made, and were soon seen in large volume. After 2 1 / 2 min- 
utes the fumes began to decrease in volume and practically ceased at 
4 minutes. 

Tests were made both with the container and without the container. 
Maximum temperatures (Fahr.) developed were as follows, as shown 
also in Fig. 1. (the first figures given are those obtained when the con- 
tainer was used) : 

Compartment 6 71- 118 

Compartment 7 1065-1065 

Compartment 8 59- 99 

In flue opposite compartment 7 600 770 

In flue top of cabinet 440- 510 

It will be noted that the use of the container materially reduced the 
temperatures in all instances but it was satisfactory to note that the 
temperatures, even without the use of the can, were greatly below the 
danger point. (See Fig. 2.) 

In view of some very valuable motion pictures placed in our cus- 
tody, we feel that the expense of a limited number of these cabinets is 
justified. Nevertheless, we realize the need for less expensive storage 
for future expansion, yet storage that would be safe from fire hazards. 

Mar., 1938] 



Consequently we have been experimenting with an alternative wherein 
a cooling agent is the protective factor rather than insulation. 

Carbon Dioxide as Cooling Agent. Our first tests were conducted 
with carbon dioxide. An enclosed cabinet, approximately 14 inches 
square and 8 feet high, holding 29 con- 
tainers stored horizontally, was used. This 
was constructed of 1-inch pine board. At 
the top two 3V2-inch horns, leading from 
a tank of carbon dioxide through a 3 /s-inch 
hose, were installed, directing the discharge 
downward into the cabinet. Three adja- 
cent reels of nitrocellulose film were used, 
the middle one being connected with a 
heater coil for ignition purposes and the 
upper and lower ones connected with ther- 
mocouples for the purpose of reading tem- 
peratures. The final tests were conducted 
at the National Bureau of Standards the 
early part of April of this year. 

A total of 15 tests were made, with varia- 
tions in the number of containers, the type 
of container, the amount of carbon dioxide, 
the kind of actuating agent used, and types 
of shelves between containers. Naturally, 
these shelves can not be solid since there 
must be a quick and easy spread of the 
cooling agent but, in some of the experi- 
ments, asbestos plates were placed under 
the containers. These plates materially 
decreased the hazard. The National Ar- 
chives container, 3 being of heavier metal, 
showed a distinct advantage over the com- 
mercial "tin" can. 

Eight pounds was the minimum of carbon dioxide used and 51 
pounds was the maximum. Although spread of fire was prevented in 
some of the tests with as little as 17 pounds of carbon dioxide, the re- 
sulting temperature in the adjacent containers was higher than that 
considered good practice for preservation purposes. Approximately 
25 pounds applied within 15 or 20 seconds after ignition is considered 
the minimum safety margin from a fire standpoint for minimum losses, 

FIG. 3. Artist's con- 
ception of "cascade" film- 
storage cabinet of The 
National Archives (not 
drawn to scale). 

310 J. G. BRADLEY [j. s. M. P. E. 

and from a preservation standpoint a greater supply is recommended. 

Test with Water as Cooling Agent. The National Archives has 
carried on independent experiments with water as a cooling agent, and 
hopes shortly to finish these tests at the National Bureau of Stand- 
ards for final measurements. 

The first of these tests involved local application of water. A 
wooden cabinet having a capacity of 9 containers stored horizontally 
was used. In the rear of the cabinet a water-pipe 1 inch in diameter 

FIG. 4. Showing progress of film fire 20 seconds after 
ignition. Note volume of smoke given off by one reel 
of nitrocellulose film. 

was placed vertically, with slits cut in the sides so that water would be 
sprayed horizontally on the top of each container. Three adjacent 
containers, each filled with nitrocellulose film, were used with empty 
containers to fill in the other spaces. A heater coil was used to ignite 
the film in the middle container but no temperature readings were 
made. In less than 5 seconds after the electric current was turned on 
smoke was visible and the water was turned on manually immediately 


Three minutes from the time of ignition the water was turned off and 
the film was examined. Only the middle reel had burned. 

The experiment was repeated several times successfully up to 17 
seconds' lapse before applying the water. After a lapse of 29 seconds, 
however, before turning on the water, one additional reel was lost. 
It should be noted that a fusible link having a melting point of 165F 
was used, the link being placed directly in the rear of the middle con- 
tainer. This link was connected to a visible weight, the dropping of 
which was a signal for turning on the water. Although this scheme 
is quite effective and has much to commend it, further experiment has 

FIG. 5. Showing condition of three adjacent reels of nitrate film after 
test with "cascade" film-storage cabinet. The ignited reel was between 
the other two. Cooling agent operated 20 seconds after ignition. 

been discontinued (for the time being, at least) in favor of other 
schemes that appear to be less complicated. 

Our final tests with water, now in progress, are being made with 
what we call a "cascade" type of shelf within an enclosed cabinet. 
This cabinet can be either portable or can be in the form of installed 
stacks separated by partitions plus a door. We definitely recom- 
mend the door, however, for at least two reasons: (1) to guard 
against the combustion's taking the form of a flame and (2) to direct 
the heat from the burning film upward to the actuating agent. 

This cabinet, as before, calls for horizontal storage and a flue in the 
rear for the escape of heat and fumes. A thermostat or sprinkler- 
head is placed in the path of this escaping heat. The water supply 
is placed directly over and leading into the cabinet. On being 



[J. S. M. P. E. 

released it falls on the first can, cascades to the sides, falls to the next 
shelf, drains toward the middle where there is a large hole, falls to the 
next can below it, and this movement of the water is repeated until 
every can in the cabinet is covered with a 
sheet of water. (Figs. 3, 4, 5, and 6.) 

One other provision is necessary, and that 
is that there must be ridges both on the top 
of the shelf and the bottom. The ridges on 
the top create troughs for the water to flow 
in, and the ridges on the bottom serve to 
prevent the lid of the can below from com- 
ing up and closing the hole. The shelves 
are gently sloping toward the center. 

Preliminary field tests with this "cascade" 
type of cabinet have been very successful 
and the design holds great promise as an 
effective and economical method of storage. 
We hope to conclude our experiments in 
this regard shortly. It is believed, how- 
ever, that in a unit cabinet holding 20 con- 
tainers, not more than 2 reels (10 per cent 
for the stack or cabinet and a negligible 
percentage for the vault) will ever be lost 
perhaps only one, if the actuating agent is 
arranged to operate within 15 to 20 seconds. 
Other Considerations. Two other funda- 
mental considerations remain in planned 
storage. The first is based upon the need 
for keeping the film at low temperatures. 
This consists of a tempering unit wherein 
the film is brought out in sealed containers 
and left to stand while the temperature 
gradually rises by induction or radiation. 
Thus condensation of moisture is prevented. This is not a compli- 
cated device and further comment seems unnecessary. 

The second consideration is what we call a re-humidifier, with which 
the moisture content of film, particularly of cellulose acetate film, 
can be restored so as to prevent brittleness. This can be accom- 
plished in several ways, but two methods are offered in illustration. 
First, an accumulator device may be used for blowing moist air 

FIG. 6. Showing an- 
other design of "cascade" 
film-storage cabinet, on 
which experiments are 
being made. 


across the surface of the film while the latter is being slowly wound. 
Again, the film may be wound upon a fluted apron and placed in an 
inclosed cabinet wherein moist air is blown across its surface. There 
are possibly other methods equally effective but whatever scheme is 
used, the surface of the film should be exposed. 

In closing permit me to recall a metaphor used in my paper "Mo- 
tion Pictures as Government Archives." 1 I referred to motion 
pictures as an awkward youth with a bad reputation with whom nice 
little boys were not supposed to play, and who occasionally went on a 
spree and smashed the furniture. Well, it looks as if this young man 
were growing up. No longer is it in order for him to sleep on the old 
sofa in the same room with Pappy and Mammy but he needs must 
have a room all his own. We simply have to plan for him. Not only 
has he outgrown all his old clothes, not only is he still as strong as an 
ox and as mean as the devil when he breaks loose, but he is altogether 
worthy of our affection and admiration. 


1 BRADLEY, J. G.: "Motion Pictures as Government Archives," J. Soc. Mot. 
Pict. Eng., XXVII (June, 1936), No. 6, p. 655. 

2 CRABTREB, J. I., AND IVES, C. E.: "The Storage of Valuable Motion Picture 
Film," J. Soc. Mot. Pict. Eng., XV (Sept., 1930), No. 3, p. 289. 

3 Report of the Committee on Film Preservation, /. Soc. Mot. Pict. Eng., 
XXVH (Aug., 1936), No. 2, p. 147. 

A short demonstration film was projected, showing burning tests of film in cabinets 
designed by The National Archives and described in the paper. 


MR, KBRSHNER: What is being done with regard to lengthening the life of 

MR. BRADLEY: Our study of preservatives, coatings, and impregnating meth- 
ods has not been completed. I understand, however, that a study of the problem 
is being made hi Hollywood under the sponsorship of the Academy Research 
Council Perhaps the Council will have some announcements to make on that 

The gases of combustion are composed of oxides of nitrogen, carbon monoxide, 
carbon dioxide, hydrogen, methane, and traces of hydrogen cyanide. 

MR. LUBCKB: Do the gases kill by suffocation, or is the action a corrosive one? 

MR. BRADLEY: It is a corrosive action, and frequently very rapid. In the 
Cleveland Clinic fire the bodies of several persons were found sitting in chairs 
without having moved. Others lived for several weeks. 

MR. ENSIGN: Was there any indication of explosion during the time the film 
was burning? 

3 14 J. G. BRADLEY (J. S. M. p. E. 

MR. BRADLEY: The gases will not generally explode unless they are con- 
fined. That is why we use vented cans. Two important considerations should 
be emphasized: (1) Protection of property in preventing a spread of the fire. 
This is accomplished by observing the principle of unit isolation, letting the 
affected unit burn out, and venting the fumes to the exterior; (2) protection of 
human life and health. We do not ask our employees to fight film fires; the risks 
are too great. Exhaust fans should be provided to clean the room of residual 
smoke, and gas masks should be kept handy for emergencies such as rescue work. 
In brief, ninety-nine per cent of the film fire control work should be done before 
the fire occurs. 

NOTE: Mr. Bradley gave an oral summary of his paper at the Fall Convention 
at New York, October 11, 1937, for the benefit of the Eastern members who did not 
attend the Spring Convention at Hollywood. After his talk, the following discussion 

MR. RICHARDSON: As I understand it, the silver that forms the photographic 
image is permanent. How long will it last? 

MR. BRADLEY: I do not know. We believe that the acetate base will last 100 
years or more if properly kept; but our experiments have not covered emulsions 
as such. 

MR. RICHARDSON: Would it not be possible to take a series of pictures on glass 
strips of convenient length? It would seem that they would last very much longer 
than on ordinary film. Another point: please extend your remarks on the in- 
jurious effects of the gases. That is an important subject in projection rooms. 

MR. CRABTREE: Film that is decomposing but not burning gives off oxides 
of nitrogen, and when these gases come into contact with moisture hi the lungs, 
nitric acid is produced, which is poisonous and very irritating. 

MR. RICHARDSON: We have had a great many film fires in which the projec- 
tionists certainly must have breathed some of the gas. Can we get any definite 
information as to how far that would not be very dangerous? 

MR. CRABTREE: If the projectionist inhales gases from burning film, that is 
not nearly so dangerous as inhaling gases from film that is rapidly decomposing 
but not burning with a flame. The film in Cleveland, of which Mr. Bradley speaks, 
was heated and gave off toxic gases, and much of the film was not burning. If 
the products of decomposition ignite, the resulting gases are not as dangerous, 
although it is not desirable to breathe them. 

MR. KENDE: In slow-burning of acetate film, are any toxic gases given off at 

MR. BRADLEY: They are quite objectionable and are not good for the health, 
but are not as poisonous as the nitrate fumes. We do not know the conditions 
under which the combustion of film will take the form of flame or the form of 
smoke. Members of the Committee witnessed a demonstration at the Bureau of 
Standards a couple of years ago in which we burned nitrate film in a can, mainly 
to test the can. In one instance the combustion took the form of a huge cloud 
of smoke, and in the next instance it took the form of flame, under apparently 
the same conditions. If we could always get the burning to take the form of 
flame, the danger would not be so great. I do not believe projectionists or anyone 

Mar., 1938] THE FlLM-vSTORAGE PROBLEM 315 

else should be asked to fight film fires. We ask our people to run and let the film 
burn out. 

MR. HOVER: There are no records showing the number who may have been 
injured by inhaling the fumes, mainly because the medical profession was not 
aware that they were so toxic. There are quite a number of records that show 
that projectionists had inhaled fumes and about two weeks later were found dead. 

I believe it is only in the last two years, due to research started in New York 
by one of the medical associations, that these conditions have been fully appreci- 
ated. There is no way of telling how many projectionists have inhaled the fumes 
and died as a result, when their deaths were blamed on heart failure and various 
other things. 

Regarding the film fire in Cleveland, I was recently informed that one of the 
casualties in that case was a man driving a car. He drove through a cloud of the 
smoke, and his car crashed into a pole less than a block away. He was dead when 
they took him out of the car, due to having inhaled the gas. 

MR. RICHARDSON: For many years I have fought for ventilation systems in 
projection rooms that would remove smoke and gases as fast as they are formed. 

MR. CRABTREE: In the event of fire, a fan could be started immediately to 
remove the decomposition fumes, but it would not be necessary to ventilate the 
room to such an extent during normal times. 

MR. GREENE: Possibly there is some need also for specific education on the 
subject among our craft. I have never heard a projectionist speak of the pos- 
sibilities of film fire but he says, characteristically, "Brother, when she pops, I am 
gone." I have known of only one who did that. There is something, I do not 
know just what it is, that impels the projectionist to stay there and fight the fire. 

MR. CRABTREE: In attempting to preserve valuable films we should not rely 
too much upon a water supply. I can imagine instances where the water supply 
might be shut off just at the time when the film was subjected to dangerous con- 
ditions. We should try to design the vaults so as to be effective even hi the ab- 
sence of water. 

MR. BRADLEY: I probably did not make myself clear on that. We should use 
insulation, and add the cooling agent as an additional feature. Insulation is 

MR. CRABTREE: Some of the fires that have happened in recent years have 
stressed the importance of three things: (1) That the vault walls should be 
sufficiently resistant, (2) that the vents should be large enough in cross-sectional 
area and no longer than necessary, since additional length incurs additional re- 
sistance, and (5) vents should be properly located with respect to their surround- 
ings. The decomposition products of the film are explosive, and if conditions are 
such as to produce a detonative mixture the explosive force is quite considerable. 
The vent size should be adequate to reduce the explosion pressure and the walls 
sufficiently strong to resist this pressure. Baffles of massive construction and of 
fire resistant material carried outward from the building roof or walls would also 
be helpful. That being the case, I do not think there would be much danger of 
the fire's being transmitted from one vault to another, as has happened in some 
of the cases mentioned. 

In Rochester we have been experimenting with a vault that would be, perhaps, 
a happy medium between a very expensive vault and the standard Underwriters' 

316 J. G. BRADLEY [j. s. M. p. E. 

vault. The cost of the National Archives vault is probably ten times that of a 
standard vault, is it not? 

MR. BRADLEY: Our cabinets for the more valuable Archival films are quite 
expensive, figured in terms of original cost, but in view of their long life we do 
not feel that the cost is out of proportion. 

MR. CRABTRBE: In the vault in question the cans have embossings on each 
side so as to provide adequate air space between adjacent cans. Air is an excellent 
heat insulator. The can is fitted with a vent so that decomposition gases can 
escape; and if the gases become blow-torches, they blow in the direction of the 
sprinkler. With such a vault, one must depend upon water. If the water supply 
did fail, however, I do not think the heat would transfer to other cans; and per- 
haps not more than one or two cans of film would be lost. This type of vault, of 
course, is not as foolproof as the one that Mr. Bradley has described, but I think 
it is much more satisfactory than the standard Underwriters' vault, which is con- 
cerned merely with the protection of property external to the vault. Apparently 
the Underwriters were not very much concerned about the film inside the vault. 

There is no doubt that if a can of film in the conventional vault does happen to 
catch fire, the water from the sprinklers will have access to the remaining cans 
and the film will be ruined. The proposed modification, we hope, will prevent 
access of water; and insure that decomposition will not progress beyond the can in 
which it originates. 

The objection to Mr. Richardson's scheme of using glass plates is, of course, 
the space that would be required and the difficulty of registering successive pic- 
tures. There would have to be registration marks on the glass slides so that they 
could be copied onto motion picture film. If you needed ten feet of film, that 
would mean 160 glass slides. I think some thought should be given to the sug- 
gestion, however. Mr. Bradley stated that the film in the safety vault may reason- 
ably be expected to last perhaps fifty or a hundred years. Methods of duplicating 
films have been improved greatly in the past few years, and assuming that we 
shall make still further progress, I think we can assume that within the next 
fifty years we shall be able to duplicate those films, let us say, 99 per cent 

MR. KENDE: Mr. Bradley has indicated that the life of acetate base films is 
definitely longer than that of the nitrate base. Of course, we all know that nitrate 
base films are the more dangerous. Is there any hope that the safety base will 
make its entrance into professional work in the near future? 

MR. CARVER: Progress is being made all the time in the manufacture of 
acetate film. At present it is not as good as nitrate film as regards quality and 
strength, that is, it is not as good for a short time. The cost is greater, and the 
demand from the trade is not very great. 

Not many actual facts are in sight to indicate that these difficulties may be 
overcome; but when we consider that nitrate film has been manufactured for a 
great many years and acetate film only for a few years, we know that improve- 
ments are bound to come. That is about as much as I can safely say. 

MR. BRADLEY: Considerable thought has been given to putting photographic 
images on material other than film. There is an experimental process in Germany 
of putting the images on aluminum. In the field of microphotography there is 
experimentation on photographing a whole book on one sheet of glass 8 X 10 


inches. But that would not be motion pictures, just still photography; that is 
why I did not bring up the subject before. 

I have a piece of nitrate film that was produced by the Eastman Kodak Com- 
pany in 1901, which makes it 36 years old. It has suffered a great deal of punish- 
ment in being carried about and handled, but it is still in almost perfect condition. 
That is why I say that even nitrate film, handled properly and cared for on the 
principle of good housekeeping, can be expected to last fifty or a hundred years. 
I regard the present form of film as basic. If we can continue our researches on 
what we have, I feel that it would be better than trying to go off into glass or 
bronze or gold or some of these other things. We have considerable promise 
that cellulose film will last several centuries. 



Summary. Correct engineering design of theater sound and projection equipment 
depends upon close contact between the engineers and the projectionists. The position 
of the projectionist in the industry is stressed, since the motion picture performance 
is placed before the audience by him. 

It is natural that there should be a Projectionists' Session at any 
important convention of the Society of Motion Picture Engineers. 
For one thing, a motion picture engineer can not be truly qualified 
in his profession if he does not know of the status of projection and of 
the problems in that field that still await solution. The engineer 
can not hope to learn of these problems from his inner consciousness 
or work out their practical answers by a feat of imagination. It is 
only by being in close contact with the projectionists who are skilled 
in the practice of their art and who daily encounter the problems in 
question that the engineer can be guided aright and can hope to pro- 
duce workable equipment and methods that will advance the art of 
projection. Solely by close and continuous cooperation between the 
skilled projectionists and engineers can apparatus be produced that 
will stand the harsh test of everyday wear and tear. 

But this is only a small part of the reason for a Projectionists' 
Session at an SMPE convention. As has been correctly pointed out 
projection of the motion picture and reproduction of the related sound 
constitute the "neck of the bottle." The motion picture industry 
has gone to enormous pains to produce an appealing and valuable 
product. Vast studios having the most modern equipment have been 
built. The most popular stories and the most successful plays are 
purchased for motion picture adaptation at figures that truly approach 
"colossal." Stars of the first magnitude are selected as the chief 
players, and at rates that are truly "awe-inspiring." Large groups 
of writers, re- writers, and specialists in the literary polishing of the 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received August 
23, 1937. 

** New York, N. Y. 



script bend their efforts to producing the best results on the set. 
Cameramen, sound recordists, electricians, make-up men, wardrobe 
mistresses, carpenters, painters, and a host of others form an army 
under the command of the most capable directors. Editors and 
cutting-room experts toil through weary months to produce the 
smoothest and most appealing film. Elaborate sales and distribu- 
tion systems carry the films to every corner of the land. A myriad 
of sparkling and attractive theaters with bright lights blazing in 
the lobbies and the names of the stars and the play in twinkling lights 
overhead attract the audience. A whole branch of the advertising 
industry is devoted to acquainting the public with the romance of 
the film presentation. Comfortably upholstered chairs, spick and 
span ushers, air-conditioned theaters, and other instances of efficient 
management are added to complete the attractiveness of the theater. 

And now we finally reach the merchandise if so prosaic a name 
may be used that it is proposed to sell. And this merchandise is 
nothing more than foot-lamberts from the screen and acoustic watts 
from the loud speakers. These are the neck of the bottle, and it is 
for these that the audience pays or does not pay. And these all 
important elements in the motion picture industry the sole and 
final reasons for its existence are under the care of the projectionists. 
If the engineers have provided proper equipment in the projection 
room, the projectionist then determines the quality and reliability of 
the performance. He is really the physical impresario of the motion 
picture presentation. Failure of equipment or incorrect handling 
can annul all that has gone before. 

The aim of the projectionist is to produce pictures that are sharper, 
steadier, and (within limits) brighter than heretofore. Much can 
be said in detail relative to each of these requirements. Nevertheless, 
broadly, they cover the field. At the same time, more perfect and 
realistic sound reproduction is also the aim of the projectionist. 

Many elements in the field of projection are in a state of evolution 
illuminating sources, projection room equipment and routine, color 
projection, and many other developments. 

It is clear enough why there should be a Projectionists' Session at 
any SMPE convention. Indeed, it would be difficult to pick any 
element in a convention that was more necessary than this session, 
which, like those that have preceded it, will contribute to the pleasure 
of future theater audiences and the prosperity of the industry that 
the engineers and projectionists alike will serve. 


Summary. Advanced methods of licensing projectionists in the Province of Al- 
berta are described, with some comments on the apparent benefits derived from the 
process. Becoming a first-class projectionist requires a licensed apprenticeship of at 
least twelve months, followed by one year as third-class and, later, one year as second- 
class projectionist before taking final examination for a first-class license. Each 
period, except apprenticeship, is preceded by a thorough examination. 

The purpose of theater regulations in the Province of Alberta 
(Canada) is to give public protection against panic or serious mishap 
in the event of film fires in projection rooms. That purpose has been 
developed to include general precautions covering width of aisles, 
door spacing, seating arrangement and fire-prevention measures 
within the auditorium, or that total enclosure that is known as a 

As film fires occur chiefly in the projection room, any measure 
that tends to confine or limit such fires to the projection room is 
considered as treating the danger at its source. This paper, however, 
will not consider details of construction, or projectors and equipment, 
but will deal with the importance of licensing and progressively 
grading the persons who work in projection rooms and who are in 
charge of projection equipment during the time the theater is open 
to the public. 

Over a period of several years Alberta has been privileged, with 
the friendly "assistance of its exhibitors and projectionists, to build 
up a system of issuing certificates to projectionists that it is believed 
represents a major factor in fire prevention. In addition, the life 
of current prints is prolonged under competent handling, and prints 
are less subject to premature mutilation than would be the case 
if novices were employed in place of the trained and experienced 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received Septem- 
ber 1, 1937. 

** Chief Inspector of Theaters, Province of Alberta, Edmonton, Alberta, 


A person wishing to take up projection work in Alberta is required 
to obtain an apprentice license. The application for such license is 
in the form of an affidavit, and must be signed by three persons, 
namely, the proposed apprentice, the manager or owner of the 
theater, and the projectionist under whom the apprenticeship is to 
be served. Thje object of this threefold preliminary is to establish 
the good faith of the application; to insure that the owner of the 
equipment has consented to let his property be used; to guarantee 
that a projectionist is willing to act as teacher or instructor ; and to 
start a genuine record of the apprenticeship so that the required 
projection room service may be readily traced when the apprentice 
is ready for his initial examination. 

The apprenticeship must cover a period of at least twelve months, 
with not less than three hundred hours of actual projection room 
instruction on standard 35-mm. equipment. Examinations are held 
quarterly, and assuming that the apprentice has fulfilled the require- 
ments as to length of time and hours of training, he may sit for the 
initial third-class examination. All examinations are written, but 
the examiner may test the ability of the prospective applicants on 
their own equipment from time to time while on inspection work in 
connection with the theater. 

The successful apprentice in the initial examination is granted a 
third-class certificate, which enables him to hold a position as assis- 
tant in small theaters and so to continue the next period of twelve 
months together with at least another five hundred hours of actual 
projection room experience. His next examination, which is for 
second-class, is advanced in both theory and practice over his initial 
attempt and is intended to represent the next higher standard at- 
tainable with the normal growth and development of the progressing 
candidate. If successful, he is given a second-class certificate, on 
which the same period of twelve months is again required but with 
not less than six hundred hours of additional projection room ex- 
perience in a second-class theater before he may sit for first-class 

Applicants coming from places outside the Province of Alberta are 
required to furnish proof of their experience as projectionists before 
examinations will be granted. This proof may be in the form of 
letters from former employers, or the original license if such license 
was required in the territory in which they claim to have been em- 

322 G. P. BARBER (j. s. M. p. E. 

Examination fees are fifteen dollars, for third-class ; twenty dollars 
for second-class; and twenty-five dollars for first-class. These fees 
cover the issuance of certificates but are forfeited should the candi- 
date fail to pass the examinations. The casual observer might be 
inclined to call these fees excessive, but a little reflection will show that 
they induce a more serious attitude of mind on the par^ of prospective 
candidates than is the case when the financial consideration is 
negligible. The examination is primarily designed to prove the 
candidates' understanding of projection room practice in its various 
phases, and a certificate is intended to certify that the holder thereof 
is familiar with the work that his employer expects him to perform. 
The fee is likely to deter the trifler, but it lends incentive to the 
serious-minded candidate in the study of real projection problems. 

Theaters are graded according to seating capacity. A place with 
five hundred or more seats requires two first-class projectionists, 
one for each projector. Two second-class men are required in 
theaters with less than five hundred but more than three hundred and 
fifty; below three hundred and fifty seats, one second-class man in 
charge with a third-class assistant. Apprentices are not employed 
in place of licensed projectionists and not more than one apprentice 
is allowed to each theater at any time. 

Certificates are issued as from January 1st of each year, and ex- 
pire on the 31st of December of the year of issue. First- and second- 
class certificates are renewable without re-examination, but holders 
of third-class certificates have the option of trying for second-class 
or rewriting for third-class, as that class is not renewable. This 
policy has been adopted as tending to urge the beginner to reach 
the higher grades. 

Question papers are changed and modernized from year to year, 
and no candidate receives the identical paper twice. Questions are 
arranged numerically under the general headings of mechanics, 
optics, electricity, and safety. There are thirty questions in the 
third-class examinations; fifty in the second-class; and seventy in 
the first-class. The percentage required to pass is 60 for third-, 
and 80 for second- and first-class. 

Should a candidate be dissatisfied with the markings of the ex- 
aminer, provision is made for a Board of Appeal, consisting of a repre- 
sentative of the Government, a representative of the exhibitors, and 
a representative of the licensed projectionists. The findings of this 
Board are binding upon both the candidate and the examiner alike; 


there is no higher appeal. While there have been many failures in 
past years, especially in the lower grades, the percentage of appeals 
has been almost nil, as the sincere candidate is fully aware of his 
shortcomings immediately the examination is completed; and as the 
whole procedure is obviously not to trick or hinder but to encourage 
and educate the applicant right from the early stages, the entire sys- 
tem builds up a spirit of confidence not only in the methods followed 
but also within the candidate himself, which makes for friendly co- 
operation among all concerned. 

Generally speaking, the system of licensing is of benefit all around. 
Those supplying films are assured that reasonable care will be taken 
of their prints, since the men to whom they are entrusted are ex- 
perienced and have "grown up" in the work of projection. Every 
projectionist must sign a complete film report for every item on the 
program (feature, comedy, and shorts) at the end of custody of each 
run of pictures. A copy of this report is sent to the Department and 
one to the film exchange. A third copy is retained for projection 
room records. We thus have accurate knowledge of the condition 
in which the film was received at any particular theater, the number 
of times it was projected, who projected it and what, if any, trouble 
was encountered. 

Exhibitors now depend upon the licensing system in selecting pro- 
jectionists, and projectionists are protected from the "fly-by-night 
machine operator" who usually may be depended upon to leave be- 
hind him a lot of trouble for the projectionist to clean up. 

It is believed that our examination fees are effective in com- 
pelling study. No candidate likes to fail, but to fail and lose money 
in addition makes the sensation doubly painful, since he not only 
gets nowhere but he foots the bill for the experience. Then there is 
the matter of three examinations, thitd-class, second-class and finally 
first-class. If he is to "arrive" at all he soon comes to the con- 
clusion that study, assimilation, and practice are essential to his 

Our biggest problem has been to combat the argument that such 
restrictions are not enforced or even suggested in other places, etc, 
etc. That is why we welcome symposiums such as are presented 
at SMPE conventions. We feel that if similar systems could be 
established throughout this whole continent there would be better 
prints, better all-round projection, and considerably less willful, 
careless, and useless waste. 

324 G. P. BARBER [J. s. M. p. E. 

It has been the wish in this paper to outline briefly a system of 
licensing that has for its main objective the purposeful conscientious 
study and application of projection knowledge on the part of those 
who spend the greater part of their working hours in projection 
rooms, and who are charged with the responsibility of putting the 
picture on the screen. These men, known as projectionists, can either 
make or mar the work of film producers and their staffs. It is 
believed, therefore, that aside from the public safety angle any 
prevention of waste in the form of mutilated prints is worthy of the 
best consideration of Governments, examiners, and the serious- 
minded thioughout the entire industry. 


MR. RICHARDSON: This paper is one of the most important that have come be- 
fore this body, and I wish to emphasize that it is a government paper. Although 
the candidates have to put up a considerable sum, the man who puts up $10, $15, 
or $25, knowing that its return is contingent upon his passing the examination, is 
not going to take much chance on passing. He gets busy and really studies. 
That is the important point of this paper. 

MR. KESSLER: For the past thirty years as a member of Local 306, and, prior 
to that, Auxiliary 35, I know it to be a fact that the examinations we formerly 
took on Park Row (New York) and the examinations that are taken today are 
given in the same old-fashioned way. Nothing has been improved. It would be 
a good idea if the engineers of today would take it upon themselves to go down and 
show the City Department how to examine projectionists. 

MR. McGuiRE: When examinations were introduced by the Government some 
years ago in Canada there was considerable opposition. They were, however, 
taken by the men who passed them with flying colors and everyone was very much 
pleased with the results. 

MR. HOVER: I believe a number of members here are familiar with the stumbling 
block of the system. If our engineers and projectionists could find a way of root- 
ing out politics from the examining boards, there would be no trouble. 

MR. FISHER: I was a member of the Board of Examiners of my city for six 
years. One day our mayor said, "We have twenty-five or thirty janitors in our 
schools; they all must have licenses to run motion picture machines." 

The corporation counsel was told by the mayor to see what he could do about 
removing from the law of the State the clause that requires that applicants serve 
apprenticeships, so that the janitors in the schools could run the motion picture 
machines because the Board of Education could not afford the cost of 

The following day I happened to go into one of the schools containing an as- 
sembly hall with five hundred little children in it. The picture machine was set 
up at the back of the hall no booth ; four or five reels of film on a chair, not even 
in tin boxes; and cable running from the machine the whole length of the hall, 
over which some little child might stumble and probably start a nice fire. I 


stopped the show, and later the principal went to the mayor's office and asked 
that I be put off the examining board because he could not run pictures that 
would endanger the lives of all those children in the assembly hall. 

I do not doubt that there are similar situations all over the state, but it is 
pretty bad that we have to put up with such things. 

MR. RICHARDSON: One bad feature of projection examinations, aside from 
politics, is that our laws take cognizance of practically only one thing, so far as 
concerns examinations, and that is fire hazard. No attention is paid to the haz- 
ard to eyesight, to the ability to put an image on the screen in such a way that 
there will be a minimum of eye-strain; and yet the shows increase in length until 
now they are three hours long. The law pays no attention to the quality of 

MR. FINN: I question the necessity for the distinction made in the paper be- 
tween, say, a 600-seat theater and a 750-seat theater, from which it would appear 
that the need for good projection is not so pressing in the former as in the latter. 

Canadian procedure makes quite a point of examining the applicant on what 
they call "his own equipment." That permits a man to be a fine projectionist in 
one theater and probably a very bad one in another house with different 

MR. GREENE : The great majority of us have probably long since reached the 
limit of patience with law-making and law-enforcing bodies; we no longer ex- 
pect anything of them, and are content if we can just keep them from doing too 
much harm. All the more it becomes the obligation of each local to assume as its 
own duty and responsibility that of safeguarding the public, and to insure the 
excellence of the performance they place before the public. They would be quite 
satisfied if through their own efforts they were able to do that without too much 
interference from politics. Fortunately there is an increasing proportion of union 
members who are not primarily interested in politics, either inside or outside their 
own organizations. 

MR. EDWARDS: About three years ago the Projection Practice Committee 
spent considerable time on an inquiry from officials in Canada regarding projec- 
tionists' examinations. I have not heard of the results of examinations taken 
directly under our suggestion, but I do believe that under the examination as sug- 
gested at the time anyone who went up for examination, first of all, could not as- 
sume that the examiner did not know his business. 

Our problem is to try to draft an examination or make suggestions for an ex- 
amination for the people who are supposed to do the examining; hi other words, 
to teach the examiners their business. Not all examiners need teaching; we 
have present a gentleman who is an inspector in Connecticut, where I know that 
the examinations are fair and strict, and the general proceeding one that should be 
welcomed by projectionists everywhere. 



T. P. HOVER** 

Summary. Engineers as a group are backward in dealing with problems involving 
the human element. They would rather deal with things than persons. They can 
not be blamed for this, however, because most engineering problems are solved by 
definite formulas and procedures, while problems dealing with the human element 
seldom follow expected paths. The human dement is a vital consideration in the 
successful operation of a theater that requires that sound and projection equipment 
be maintained in first-class condition at all times. 

Plans and ideas that have aided in maintaining high standards of projection 
in Lima (Ohio) are presented. Since the city is more than 150 miles from the nearest 
parts-supply company, a well-planned system of mutual cooperation is of the greatest 
importance in order to prevent shut-downs with attendant loss of money and good- 
will. Success over a period of ten years recommends the plans to the consideration 
of other protectionists' organizations that are more or less isolated from repair and 
emergency engineering facilities. 

Many changes have occurred in the motion picture industry since 
the arrival of the first practicable projectors. However, two out- 
standing objectives remain unchanged and are of vital interest to 
the exhibition branch of the industry. The most important is that 
projection equipment be maintained at the very highest efficiency at 
all times. Almost of equal importance is the fact that such main- 
tenance should be carried on at a price within reason. 

Methods of solving these problems when motion pictures were in 
their infancy differed somewhat from those of the present day. A 
breakdown of the single projector meant long weeks of waiting for 
repair parts, which only too often did not even resemble the originals 
and sometimes never did arrive. Manufacturers of equipment were 
not entirely to blame, for if they attempted to "sell" the idea of 
parts or repair service, the suggestion was often greeted with the 
attitude that perhaps the machine was not so good in the first place 
if provision had to be made for its repair. 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received November 
7, 1937. 

** Warner's Ohio Theater, Luna, Ohio. 



The simplicity of design and construction of many of the older 
types of projectors often assisted the ingenious projectionist. In 
1926 the author was called upon to check over a Powers projector 
to which had been applied successfully repair parts from a lawn 
mower, a sewing machine, a gas engine, an alarm clock, and a re- 
volver. The great precision of modern equipment has definitely 
obviated these haphazard sources of repair parts. High-speed 
communication and transportation have been of great assistance to 
the exhibitor. Standardization of parts and whole-hearted co- 
operation from equipment manufacturers have also been helpful 

One of the most serious causes of annoyance and loss of revenue to 
the exhibitor is the class of trouble broadly referred to as "break- 
down." No equipment is free from the likelihood of breakdown, 
and the only remedy is a maintenance plan that will reduce the 
probability of its occurring and a carefully planned repair program 
that will bring back into operation the affected equipment with the 
least expenditure of time and money. So much has been said and 
done on the subject of maintenance and service that they will not 
be discussed here further, but the problem of emergency repairs will 
be given considerable attention. 

Approximately fifty cities in the United States have supply depots 
and theatrical repair shops giving 24-hour service. Emergency 
repairs in those cities resolve themselves largely into procuring and 
installing damaged parts, which can usually be done in less than an 
hour. For theaters located even short distances from repair facili- 
ties, an entirely different problem appears. Telephone or telegraph 
may instantly connect the exhibitor with the nearest available 
source of repairs, but the transportation of the parts, which even at 
best may take from four to twenty-four hours, is an important item. 
A solution of this problem has been worked out in Lima (Ohio) by 
the projectionist members of the local Union. The average city 
of any fair size has facilities within it for making almost any repair 
that may be necessary to keep the picture on the screen until the 
proper parts arrive from the factory or a serviceman can bring them. 
In order to be available at a moment's notice, these emergency 
facilities must be properly recognized and coordinated. The key- 
note of the system must be cooperation between the exhibitor, the 
projectionist, and the holder of such facilities. The projectionist 
who will devote a small part of his time to "selling" his fellow towns- 

328 T. P. HOVER [j. s. M. P. E. 

men on the idea that his work of projecting pictures is a profession 
entailing scientific knowledge and accuracy will find that he has 
driven an entering wedge into his problem. 

The most important prerequisite of emergency service is the pos- 
session of the proper testing equipment and tools. Through care- 
ful planning within our organization, members seldom buy equip- 
ment of types already held by other members. Ten members of an 
organization each buying a $100 amplifier analyzer and test kit, 
will have available only $100 worth of test apparatus in the event of 
an emergency; but the same ten members, with proper planning, 
can have available a full $1000 worth of modern equipment ranging 
from tube-tester, oscilloscope, vacuum-tube voltmeters, all the way 
to complicated vibration analyzers. The fact that some of the larger 
chains have made available to their engineers a quantity of labora- 
tory testing equipment is ample proof that the investment is justi- 
fied. Our own group, through careful buying, has available almost 
$5000 worth of precision equipment. Included in the equipment 
are a number of complete portable amplifiers, which, in the event 
of a breakdown could be immediately hooked up to any projection 
room in the city. It should not be thought that this equipment has 
been purchased for exclusive use in projection rooms, for the income 
available from such equipment when so used would not justify the 
expenditure. For instance, the projectionist may add to his income 
by renting out public address systems, which have been constructed 
to handle practically any photocell input. Such equipment ob- 
viously, is available also for emergency service work. The avail- 
ability of this equipment means that any amplifier or loud speaker 
breakdown can be completely remedied in less than thirty minutes. 

In the event of minor amplifier repairs, such as of defective re- 
sistors or condensers, a local radio parts company provides day or 
night service in return for our permitting them to mention in their 
advertisements the theaters that use their brands of vacuum tubes. 
Those who have attempted to arouse a parts supplier at night in 
order to get an elusive amplifier part, can readily appreciate the im- 
portance of reliable day and night service. 

Mechanical troubles offer the possibility of the most extended 
shut-downs. While a reasonable quantity of spare projector parts 
is usually carried, breakdowns often occur in which gears are damaged 
or shafts bent in such a way that factory replacement is almost 
necessary. Records of ten years of mechanical breakdowns have 



shown that 95 per cent of such breakdowns are the result of using 
cheap, mismatched, or so-called bootleg repair parts. 

Our organization has effected a cooperative tie-up with a local 
company specializing in the manufacture and repair of motors. All 
regular motor repair work of the theaters is done by this one company. 
In return, a machine and electrical shop that is a veritable me- 
chanic's paradise is open to the use of our members. This company 
also maintains a research department and has portable welding and 
brazing equipment and a portable power and lighting unit, which is 

FIG. 1. 

Experimental projection room maintained by local pro- 

also at the disposal of our members. As the shop runs 24 hours a 
day, its facilities are always available. 

Recently the main drive gear of one of the projectors was ripped 
out due to the "freezing" of a misfit shutter shaft. A telephone call 
revealed that to obtain the nearest replacement part would require 
an automobile trip of almost eight hours, or high-speed railroad 
shipment of six hours. The breakdown occurred at 5:30 P.M., and 
one projector continued the show. A collection of gears from ob- 
solete miscellaneous projection equipment was assembled and taken 
to the machine shop. Two and a half hours later, after multiple 
grinding, brazing, turning, and boring operations, the disabled 
projector was again in service. The replacement gear was so nearly 

330 T. P. HOVER [j. s. M. P. E. 

identical to the original that the new gear, which subsequently ar- 
rived, has been held as a spare, and the emergency gear has never 
been removed. A point of interest is that the entire cost incident to 
rebuilding the gear was less than the factory cost of the new one. 
Almost every projectionist envisions a projection room where he 
can try out new ideas, uninterrupted by the routine of theater 
operation. Twenty years ago our organization moved to make this 
dream a reality. Sponsored by the local high-school principal and 
the student council, members of our group designed and supervised 
the construction of a modern projection room in the high-school 
auditorium. Under the supervision of the author and Professors 

FIG. 2. Screen and amplifier in local high-school audi- 
torium, supervised by local projectionists. 

R. E. Offenhauer and H. W. Leach, visual instruction pioneers, this 
projection equipment has played an important part in the educational 
program of the high-school system. The equipment and the audi- 
torium have been available to our membership at any time. An 
elaborate physics and chemistry laboratory, and an industrial arts 
department, as well as a fair technical library, are also open to our 

While the depression slowed up our activities, the interest of the 
student body has been responsible for the installation of a sound 
system which has been completed for use during the present year. 
This system was entirely constructed by projectionists and is one of 
the most elaborate in any school in the country. Fig. 1 shows the 
projectors and the photocell amplifier used in this installation. 



Voice reinforcement, radio and non-synchronous equipment are 
available. Hard-of -hearing aids are being installed for the benefit 
of those so afflicted. Fig. 2 shows the auditorium stage with the 
microphones, speakers, and amplifier cabinet located in the or- 
chestra pit. It would be almost impossible to list all the original 
ideas that have been developed in this projection room, and special 
work has been conducted in practically every branch of projection. 
One of the most important benefits derived from our cooperative 
tie-up with the school system is that the students, and through them 
their parents, realize that our projectionists' organization is an 

FIG. 3. 

Demonstration equipment for lecture: 
Movies Their Voice." 

"Giving the 

important civic asset. Our members are welcomed as guest in- 
structors in both the public and parochial schools. Their knowledge 
of the physical sciences permits them to aid the teachers in present- 
ing practical experiments and demonstrations. This works a two- 
fold benefit, creating interest in our profession among students and 
teachers, and stimulating interest in modern education among our 
own members. 

A contact committee carefully handles matters of publicity 
dealing with projection. This committee makes no attempt to pro- 
mote either theaters or pictures, and never permits itself to be used 
for ballyhoo purposes. Its activities are apparently without end. 
Physicians and surgeons request their assistance in enlarging x-ray 

332 T. P. HOVER tf. s. M. p. E. 

films and microscope slides by projection. A breakdown in the 
local broadcasting station brought a quick call for assistance. The 
police department has contacted us for the use of ultraviolet equip- 
ment, and a number of churches and schools keep 16-mm. film 
records of their most important activities. The camera work is 
always assigned to our members. We find that our offers of co- 
operation and assistance usually surprise our fellow townsmen. 
Too often the policy of exhibitors has been to contact fellow busi- 
ness men only when they have wanted something. 

Recently a prominent industrialist suggested that there was 
probably considerable romance behind the scenes in projecting 
pictures, particularly since the subject has always been a closed book 
to the public. As a result of the idea we assembled and staged a 
series of demonstrations entitled "Giving the. Movies a Voice." 
Fig. 3 shows some of the sound equipment used in this lecture. 
Originally intended for the benefit of one of our luncheon clubs, the 
demonstration has been given in three local high-school auditoriums 
and before many service and luncheon clubs. Its popularity has 
brought many attempts to book the demonstration as a standard 
lyceum act. As a builder of good will, we consider it invaluable. 
The main idea that we have attempted to carry out in our organi- 
zation is to give 100 per cent projection service to the theaters and, 
by selling our organization to the city at large, to gain their co- 
operation wherever we may need it. 

Our service activities do not represent a commercial or money- 
making project. We have no wish or intention to operate in com- 
petition to regularly organized service, supply, and repair agencies. 
Our activities are planned for the express purpose of keeping the 
picture on the screen as far as possible regardless of equipment and 
current supply failures, and to do so at reasonable cost. The fol- 
lowing case histories are ample proof of the time saved by our emer- 
gency service system. 

Case Trouble Off Sceen Nearest Service 

1 Main drive gear stripped 8 min. 4 hours 

2 Speaker field burned out 25 min. None 

3 Generator completely disabled 22 min. 10 hours 

4 Motor control transformer burned out 2 min. 6 hours 

5 Tubes in voltage amplifier burned out by 

current surge 100 min. 6 hours 

6 Film damage due to worn parts (Hopeless) 


Our service facilities are not limited to this city alone, but are 
also available to projectionists in the many nearby small towns. 
However, managers in these towns who will not pay reasonable 
wages to their projectionists are politely informed that our assis- 
tance is limited to exhibitors who are willing to cooperate with their 
employees, and that a reasonable compensation to their projec- 
tionists is the surest indication of such a spirit of cooperation. 

The author wishes to acknowledge the helpful criticism of F. H. 
Richardson in connection with the preparation of this paper. 


MR. RICHARDSON : I should like to call your attention to the improvement that 
is gradually occurring in the class of men now engaged in projection. A few years 
ago there were few who could be induced to study the principles of projection. 
Some time ago I had the privilege of addressing the Lima Local Union on pro- 
jection matters, to which meeting were invited members of unions of other cities 
and towns. Many of these unions were represented, although in some instances 
the round trip was more than 100 miles, which indicates a very encouraging 

MR. EDWARDS: In speaking of managers who are generous enough to buy a 
spare tube and lock it in the safe, Mr. Hover assumes that that is small town prac- 
tice. We have theaters in New York in which six to eight carbons are taken out 
of the safe in the morning and handed to the projectionist, who has to make them 
do for the whole day or he is out of luck. 

One of the points where we need cooperation most is between the managerial 
staffs and the projection rooms. The cooperation is generally all from the one 
side, especially as regards material. On several occasions I have tried to induce 
the Society, through the Projection Practice Committee, to try to do something 
about educating the managers, but to date we have not gone very far with it. 

In the majority of theaters the manager has no technical interest in projection. 
He takes projection as a matter of course, just as we do when you press a button 
to light a room in the home. It does not occur that in pressing the button we 
utilize hundreds of different agencies to produce that light. All we think about 
is that we press a button and the room lights up. Managers very largely do the 
same thing; their chief interest is on the lobby and the house. 

MR. HOVER : To work a plan such as ours, in a small town, we must have the 
cooperation of the exhibitor, if only to the extent of placing a few passes appro- 
priately. Our system has not solved the service problem. All it has done is to 
increase the speed with which we can get the picture back on the screen in case of 
stoppage. Success has been largely due to the fact that, with one exception, and 
that an independent house, we have had wonderful cooperation from the managers. 

It is only fair to say, however, that the sound and servicing system in most of 
our theaters is very good. It is a fact, of course, that a breakdown usually oc- 
curs when the service engineer is elsewhere that seldom fails; but during a pe- 
riod of five years, the engineer has never been called into a theater to find the pic- 
ture off the screen. The show was always running long before he arrived. 


Summary. An argument is presented for recommended standard screen-image 
proportions based upon the dimensions and brightness that will provide most com- 
fortable viewing conditions at the center of the theater seating space. It proposes 
that such compromise would tend toward best average viewing conditions and, therefore, 
least eye-strain and most enjoyment for the audience as a whole. 

During past years various bodies have discussed screen-image 
dimensions from various points of view. Some have put forward 
recommendations as to what conditions might be regarded as correct 
for various seating arrangements. There does not appear to have 
been any point-by-point comparison of large and small screen images, 
with relation to the observer, upon which reasonable opinions as to 
the desirable shape and size may be formed by the exhibitor, the 
theater manager, or the projectionist. 

For the theater audience the size of the image has a bearing on the 
ease of viewing the picture, freedom from visual weariness or eye- 
strain, and degree of enjoyment of the picture. The problem is 
quite complex, for the reason that from no two seats in any theater is 
the view of the screen the same ; although it may reasonably be held 
that over a considerable area of the auditorium the conditions may 
not vary to any great extent. However, in other sections of the 
seating area there are very appreciable differences, often of such 
extent as to abuse considerably the eyes of a portion of the audience, 
especially if the dimensions of the screen image are not happily 
chosen. In average theaters the best we may hope for is greatest 
possible visual comfort for a portion of the audience, with as little 
abuse as possible for the remainder. 

First of all, as is obvious, the size of image that would provide 
greatest eye comfort in a long auditorium for patrons seated at the 
front would be quite inadequate for comfortable viewing by those in 
the rear rows of seats. Were screen illumination kept at such value 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received August 
16, 1937. 

** Quigley Publishing Co., Inc., New York, N. Y. 


as to afford greatest comfort for those occupying the front seats, it 
would not be sufficient to enable patrons at the rear to distinguish 
details of the picture without straining their eyes. 

On the other hand, if, by virtue of either high illumination or great 
magnification, patrons seated, say, 125 feet from the screen were 
able to view the picture with ease, a poor condition would exist for 
those at the front. Either the illumination would be too great, thus 
inducing eye-strain; or the eyes would be forced to cover a wide area 
in following moving objects over the screen. Moreover, the image 
would be greatly distorted for those seated in the side front seats. 

Moving the eyes to follow action over a large screen at short view- 
ing distance may not be very important, if done intermittently or for 
only short periods of time. But performances nowadays last from 
two to three hours, with practically no relief; whence it should be 
evident that screen-image dimensions that are unnecessarily large 
should be avoided. 

Another point to be considered is that as the size of the image in- 
creases, the distance at which the image appears clear and sharply 
defined increases also, and it seems to be important that the Society 
should not only establish the relation between maximum permissible 
magnification and front-row viewing distance, but should also im- 
press upon exhibitors and theater managers the importance of not 
exceeding the limits so established. There are theaters even today 
in which the magnification is so great that the image is not acceptably 
sharp as far back as the tenth row of seats. 

What is the criterion for correct screen-image size? The question 
can not at the present time be answered definitely, and before an 
answer should be even attempted, certain other factors must be ex- 
amined carefully. 

Ability to see is a complex of brightness and size of the object, and 
the distance from which it is viewed. If an object not brightly 
illuminated be viewed from a given distance, its finer details may not 
be clearly discernible. If the illumination be increased, more and 
more of the details will appear. But if the object be sufficiently 
illuminated as to appear at its best to an observer at the given dis- 
tance, it will be too bright for comfortable viewing, over a space of 
time, for one situated appreciably nearer. If, however, the illumina- 
tion remain as originally, but the object be magnified, the finer 
details will be more easily visible at the greater distance, but not so 
acceptable at the shorter one. 

336 F. H. RICHARDSON [j. s. M. P. E. 

The matter is consequently one of compromise between the ex- 
cellence of the view and the ease of viewing at the long and short 
viewing distances. Unfortunately, there are theaters in which the 
rear seats are 150 feet from the screen and the front ones forty feet 
or less. Some screens with images 23 to 25 feet wide are located as 
close as 15 feet to the front row of seats, and not more than 75 feet 
from the rear seats. Such extremes can not be served simultaneously. 
With regard to screen illumination, perhaps the best we can do is to 
compromise on the value that will best serve the center of the audi- 

Obviously, absurd conditions such as have been described must 
have their effect upon the box-office. Those who are in the front 
rows, within 25 feet of a 28-ft. screen-image, can see hardly more than 
an unbeautiful smudge, which does not become acceptably sharp 
until the viewing distance is increased to about forty feet. Patrons 
who unwittingly, or perhaps because no other seats are available, 
view the screen under such conditions, must leave the theater with 
strained or fatigued eyes, and the question may definitely be raised 
as to whether those patrons are likely to patronize the theater again 
unless they can be assured of seats more advantageously located. 

There are other objections to large screen-images. A larger image 
demands more light through the projector aperture to maintain the 
same brightness or illumination per unit of area. Doubling the 
linear dimensions of a screen quadruples the area, and likewise the 
requisite amount of light. The electric power demand becomes 
greater; the difficulty of handling the arc and maintaining uniformity 
of illumination is increased; and the temperature at the film-gate, 
where the film passes the aperture, is increased. The latter fact 
increases the fire hazard and dries out the film, increases the likeli- 
hood of buckling, and renders the film more brittle and susceptible to 

Some persons favor the large image with respect to improving the 
visibility of details from the rear seats, especially in deep houses. 
Others seem to prefer a smaller image, brightly illuminated. A small 
screen-image may not be appropriate in a large auditorium, but the 
size should, on the other hand, not become excessive. Small images 
of the actors may seem out of proportion to the surroundings, and 
the actors may look like pygmies; conversely, excessively large 
images may appear gigantesque. We are accustomed to judge the 
sizes of objects by comparison with surrounding objects. If the 


screen-image of a man appears three feet tall, and if the surrounding 
objects are in the same proportion, the eye will see nothing wrong. 
Furthermore, the patrons become absorbed in the story and the 
picture, and become more or less unconscious of objects beyond the 
screen while projection is in progress. 

In conclusion, it should be emphasized once again that it is im- 
portant that the Society study this problem further, and establish 
a relation between screen sizes and viewing distances and angles. 


MR. GOLDSMITH: The considerations advanced by Mr. Richardson merit study 
and certain practical conclusions. It is not possible to project upon the screen 
more detail than exists on the film, which fact is sometimes forgotten in projecting 
unduly large pictures on screens too close to a major portion of the audience. 
Present-day positive prints have a finely granular structure. Beyond a certain 
enlargement, the projected picture, if viewed fairly closely, shows crudity of out- 
line and "crawling" (or grain-motion effect). If the picture is not sharply focused 
in part (as frequently happens because of the limited depth of focus of the camera 
objective), the effect of closely viewing a greatly enlarged picture is even less 

Probably an appropriate viewing distance for most of the audience in preferred 
positions lies between four and seven times the screen width; the effect of picture 
enlargement outside this range or of viewing the picture at a lesser distance being 
disadvantageous. Discretion in selecting picture size for a given theater is there- 
fore necessary, for bigger pictures are not necessarily better pictures and may be 
distinctly the contrary. 

MR. BRYANT: I do not think that a standard adopted by this Society would at 
all times work out satisfactorily throughout the world. In several theaters we 
have built in foreign countries, we have tried to maintain projection size in pro- 
portion to the viewing distance, but it has been impossible to do so because the 
patrons are not satisfied. 

In South American theaters having 100- to 120-ft. viewing distances from the rear 
seats, we have attempted to use screens 18 to 20 ft. wide, thinking they were 
large enough if anything, too large. Within three months after the opening of 
the theaters, we had to reconstruct our screen frames and install new screens to 
give the public the 24-ft. picture they demanded, by letter and memorandum to 
the management of the theaters. We had more than 150 requests in one theater 
to increase the picture size to 24 ft., despite the fact that the front row of seats 
was within 20 ft. of the screen, due to the fact that it was a small picture house. 

MR. FARNHAM: Mr. Richardson pointed his remarks almost entirely at the 
theater owner. The architect should come in for his share of the criticism, be- 
cause often the screen is built to fit into the proscenium arch. 

A number of years ago we made some tests that indicated that the minimum de- 
sirable size of screen was very roughly one-seventh of the maximum viewing dis- 
tance. There are, of course, many factors that may disturb that ratio grain 
size and so forth; but the 7 to 1 ratio seemed to work out fairly well. This limit 


was based upon the fact that when the maximum viewing distance exceeds seven 
times the picture width, details in the picture became so small as to be seen only 
with difficulty. 

MR. KELLOGG: This problem ties in very closely with the analysis Mr. Schlan- 
ger has been making with regard to proper utilization of the auditorium seating 
area. I do not see why theaters can not be designed so as to have larger seating 
capacity within reasonably satisfactory angles, and a smaller ratio of maximum 
to minimum viewing distance, merely by eliminating a great many of the front 
seats, using more or deeper balconies than at present, with greater total height at 
the back, and placing the screen so that the patrons in the bottom level seats will 
look up toward the screen at the maximum angle of comfort. 

MR. SCHLANGER: The Projection Practice Committee has been doing a great 
deal of work toward solving this problem, and hundreds of our survey charts 
have been received from theaters all over the country, analyzing this very situa- 
tion. As Mr. Farnham says, the architects are at fault. They build theaters to 
suit themselves, in any shapes that happen to please their minds, or according to 
the real estate they happen to buy. No consideration at all is given as to what 
shape of theater will provide the greatest number of satisfactory seats within the 
limitations of the projection system and the film. It is only now that we are be- 
ginning to realize that the building must be considered before we think of the 

Mr. Farnham mentioned a ratio of seven times the width of the screen for the 
maximum viewing distance. The survey charts show an average ratio of 5.3, 
with a good proportion of the theaters far below 7. There are at least two factors 
to be considered in relation to screen size: One is the technical factor (film size, 
graininess, optical systems, etc.). The other, as has been mentioned, is the hu- 
man element, dramatic value. Those, unfortunately, usually conflict. The only 
way hi which we can tell how the two factors are related is by such a survey. The 
survey charts show the actual tendencies, taking into consideration the human 


Summary. Certain faults of perforated screens are discussed, particularly with 
relation to the perforations. The question is raised as to what extent the faulty perfora- 
tions, as illustrated, may be detrimental to sound and picture quality in theaters. 

Since the inception of talking pictures, the loud speakers have been 
placed almost universally behind the screens, into which thousands of 
small holes have been punched in order to permit passage of the 
sound-waves through the screen. The writer has a number of times 
taken issue with this practice, believing that as good acoustic charac- 
teristics, if not better, could be attained with the loud speakers 
located at the bottom, top, or sides of the screen, where structural 
conditions permit; and, by using solid, instead of perforated screens, 
the brightness of the image would be enhanced to the extent of 
approximately 10 per cent of the light lost in the perforations. 

The loss of light has its effect not only upon the brightness of the 
image, but upon the cost of power ; and the presence of the perfora- 
tions adds to the maintenance cost because of the more rapid accumu- 
lation of dirt due to the perforations. Perforated screens become 
soiled much more rapidly than do solid screens. The continuous 
currents of air through the perforations draw with them dust and 
grime that deposits not only upon the inside walls of the perforations 
but also upon areas of the screen surface surrounding the perfora- 
tions. Chemical action may be set up by these deposits that will 
gradually cause discoloration of the screen and render cleaning futile. 
The deposit, furthermore, tends to reduce the effective areas of the 
perforations, with a consequent effect upon the acoustic characteristic 
of the screen. 

If we must continue to use screens that are perforated, and continue 
to suffer the loss of light and the detriment to quality, we should be 
sure that the perforations are always clean, free, and open. Ex- 

*Presented at the Fall, 1937, Meeting at New York, N. Y. ; received September 
13, 1937. 

**Quigley Publications, New York, N. Y. 


340 F. H. RICHARDSON [J. S. M. p. E. 

amination of a number of commercial screens under a magnifying 
glass fails to show that even one of the screens studied has uniformly 
clean-cut, open perforations, although in most of the screens there is 
a certain proportion not subject to severe criticism. In some of the 
screens that are now being marketed at high prices the condition of 
the perforations is quite bad. Fig. 1 shows photographic enlarge- 
ments of perforations of screens now being marketed. 

FIG. 1. Photographic enlargements of perforations. 

The purpose of this paper is to bring this matter of faulty perfora- 
tions to the attention of the Society, to show what variability may 
be expected from screen to screen, both as regards acoustic properties 
and image brightness, and to point out the obstructions in the per- 
foration passages that lead to the rapid accumulation of dirt and the 
deterioration of the screen. 



MR. MALMUTH: Briefly, the problem can not be solved by any known means, 
because the perforating dies are subject to the ordinary phenomenon of wearing. 
To make perfectly clean perforations at all times, we should have to sharpen our 
dies almost daily. It also happens, in all cases of screen manufacture, to my 
knowledge, that the perforating is not done by the screen manufacturer. There 
are two organizations in this country that do practically all the perforating for all 
the screen manufacturers. There is no possible means by which we can adequately 
control perforations so far as to be sure they are clean-cut. 

The sound in most instances restricts the screen manufacturer to a thick- 
ness not greater than 0.015 inch, and a weight of not more than ! 3 /s ounces per 
square-foot, so that we must use what we call a 9 per cent perforation, or a hole 
that is 0.049 inch in diameter. In time the dies wear, until the holes become 
0.051 inch in diameter and larger, and then they are discarded. When a very 
small pin strikes into fabric that is only 0.015 inch thick or thereabouts, the fabric 
gives and we do not get a true perforating action; we get a tearing action. Of 
course, most screens are made of cloth or fabric and coated. It is almost impos- 
sible in a fabric screen to avoid fraying or tearing at the edges when using a per- 
forating die of the type used today. 

MR. RICHARDSON: Is it not a fact that the thicker and softer the material, the 
greater is the tearing at the edges of the hole? 

MR. MALMUTH: Most likely, but as the material is restricted to a 0.015-inch 
thickness, the effect will be about the same no matter what the surface is. 

MR. RICHARDSON: There are some screens put out that are twice as thick as 

MR. MALMUTH: Not perforated screens. There are woven screens that are 
much heavier than the requirements call for, but they are beaded. The beading 
fills the holes. Unless some other means of perforating is found or some other 
material is used for the manufacture of screens, we shall always have this 


Summary. Because commercial motion pictures on 16 -mm. film are an out- 
growth of "home movies,'' the standards of projection are low. Less care is given to 
proper presentation than in theatrical showings of 35-mm. film; whereas, because of 
greater overall magnification, greater care should be taken. Some of the more glaring 
faults are treated in detail, a general treatment is set forth, and the importance of 
proper presentation is clarified by comparison of show-windows of the street and of 
the screen. 

In the rapid expansion of 16-mm. film into the commercial and 
advertising fields there has been a most lamentable lack of attention 
to the problem of achieving optimal results in projection. When the 
16-mm. field was "growing up" the usual alibi, after demonstrations 
of highly questionable success, demonstrations "ballyhooed" as prov- 
ing 16-mm. pictures the equal of 35-mm., was, "It's not so bad for 

Now that 16-mm. pictures have been greatly improved, and ap- 
parently have retired to the field they had honorably won for them- 
selves, one can appraise the 16-mm. technic in its own conceded 
field and find plenty lacking. 

This is the natural result of the 16-mm. film's being an outgrowth 
of the "home movies" and being sold primarily on the basis of sim- 
plicity. Never has an opportunity been missed to stress the claim 
that 16-mm. operation is childishly simple, and that is why the results 
have been, for the most part, simply childish. 

To take up but a few typical items, consider first that of foreign 
matter in the aperture. The 16-mm. picture usually subtends an 
angle at the viewer's eye at least equal to that subtended by the 35-mm. 
picture in the theater, and since the 35-mm. aperture is four and 
one-half times as large as the 16-mm. aperture, the effective magnifica- 
tion of foreign matter in the 16-mm. aperture is 4 J /2 times as great as 
it is in the 35-mm. aperture. No more need be said as to the need for 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received Septem- 
ber 1, 1937. 

** Orpheum Theater, Minneapolis, Minn. 


cleanliness, yet in the field of 16-mm. projection but scant attention 
is paid to it. The 16-mm. "operator," in childish simplicity, will go 
blithely on reel after reel with a piece of dirt in his 16-mm. aperture, 
to remove which from a 35-mm. aperture the professional projection- 
ist in the theater would work like a beaver. Here is one phase of 16- 
mm. projection requiring the utmost vigilance over every source from 
which the film can pick up dirt, such as film containers, reels, work 
tables, etc. So simple a thing as picking up a reel that has lain upon a 
desk for a day or so and winding film upon it without first wiping it 
with a lintless rag or a chamois will almost certainly result in a dirty 
aperture. A further and permanent effect of this sort of neglect of 
detail will be scratches and "rain" marks on the film, and here again 
the greater magnification will show as glaring faults on the 16-mm. 
screen marks almost microscopic, which might pass unnoticed in a 
35-mm. picture. 

An excellent example of the tenacity with which dirt, once intro- 
duced into a system, will remain is afforded by the dirty aperture 
edges showing in many present-day release prints. This started with 
the studio strike of a number of years ago. Something happened to 
Hollywood at that time from which, with all its technical excellence 
and skill, it has not yet recovered. 

There may, however, be a silver lining to this cloud. Remember 
how hundreds of physicians and dentists took up photography as a 
hobby with miniature cameras, processed their first films in the same 
sloppy manner as they did their x-ray films, and were horrified to see 
the results in 10 or 12 X enlargements; how they reentered the photo- 
graphic kindergarten, learned how to process film properly, and then 
carried their new knowledge and skill back into their x-ray work, to 
its great, lasting, and much needed improvement. So, perhaps, will 
the distribution and exhibition branches of the 35-mm. section of the 
industry learn at last how 35-mm. film should be handled and cared 
for, by seeing upon 16-mm. screens the results of the faults in their 
present technic. If this happens 16-mm. experience will have con- 
ferred a genuine benefit upon the entire industry. 

The aperture plates of all 16-mm. projectors should be easily re- 
movable for cleaning on both sides and behind the plate. Such safe- 
guards are a part of the best theater projection practice. They need 
to be applied 4V2-fold to 16-mm. practice, not forgotten entirely. 
Would it be so impracticable to make the 16-mm. aperture an optical 
image like the scanning slit in modern sound-heads? 

344 C. L. GREENE [j. s. M. p. E. 

In spite of precautions the aperture-edge image will seldom, if 
ever, be clean and sharp, which renders screen masking all the more 
desirable; yet, not one 16-mm. picture in a hundred is masked. 
Again the technic is already well developed in the theater and needs 
only to be applied in this field. 

In the same manner may be treated the matter of control of ex- 
traneous light. None that is perceptible should reach either the 
screen or the eye. The details have long since been worked out, and 
are so well known that it would be pointless to repeat them here. 

To those who know the elaborate precautions professional projec- 
tionists take to keep their projector optical systems free from oil 
and dust, the questionable care given 16-mm. optical systems is rather 
shocking, and is further aggravated by the fact that most 16-mm. 
projectors are first-rate oil throwers. Certainly their optical systems 
can be kept in fit condition to achieve the magnification and resolution 
demanded of them, but more knowledge, care, skill, and work are re- 
quired than often goes into the commercial showing of 16-mm. pic- 
tures. The old alibi "it's only 16-mm." is still in use, and still covers 
a multitude of sins sins not always of the equipment. It would help 
considerably, however, if all manufacturers would so build their pro- 
jectors that all elements in the optical train were either readily and 
completely accessible for cleaning or else readily removable, and 
those that are removable should be replaceable in adjustment,; and, 
in the case of lenses in focus, something akin to the lens-mounts in 
theater projectors. 

Into the commercial field, where commercial motion pictures with 
their own set of high and rigorous standards ought to be, "home 
movies" have come, bringing with them their own lamentable lack of 
standards. The change from 35-mm. to 16-mm. film does not auto- 
matically put all projection troubles into the discard. Rather it ushers 
in a whole family of new ones, faults that in a theater would send a 
patron post-haste to the box-office to demand a refund, but which we 
instead needlessly tolerate because "it is only 16-mm." The optimal 
presentation of 16-mm. film does not call for less knowledge, skill, care, 
and work than does 35-mm. ; rather it calls for more. A 35-mm. pro- 
jector and film neglected to the same extent as the average 16-mm. 
projector will show a superior screen-image. 

In the word work is wrapped up much of the present difference 
between commercial 16-mm. projection and theatrical 35-mm. pro- 
jection. In the former the fixed quantity is the amount of work the 

Mar., 1938] 16-MM. PROJECTION FAULTS 345 

projectionist is willing to do ; the dependent variable is the quality 
of projection that that much work will produce. In theatrical show- 
ings the screen-image quality is the fixed quantity, and the dependent 
variable is the amount of work necessary to produce that quality. 
Not long ago the writer overheard a conversation between a profes- 
sional projectionist and a would-be one that well illustrates the point. 
The projector was a cheap one, and due to improper fit of the film 
in the upper edge-guiding assembly there was a quiver in the picture 
that the professional was suppressing with finger pressure. In a tone 
of surprise the other exclaimed, "But you're not going to hold your 
finger there all night." The reply was, "I'm not going to have my 
picture quivering all night." Two diametrically opposite points of 
view, two different conceptions of the dependent variable, they illus- 
trate much of what commercial 16-mm. projection deserves, yet 

The only advantages of 16-mm. pictures are reduction of bulk, 
weight, and cost. These are indisputable and powerful arguments 
for its adoption, but it should not be adopted under the delusion that 
it is childishly simple. Whenever it is so adopted the results are 
simply childish. 

No good merchandiser ever gives a moment's consideration to any 
but the highest possible standards in dressing his show-windows, and 
with all his care it is only rarely that he creates a window that he 
would deem worthy of preservation as a still photograph. Every 
motion picture made and used for advertising purposes is a show window 
that was deemed worth photographing. It is one of those "once in a 
blue moon" achievements, and the projection of the film is the re- 
creation of the show-window with this additional condition imposed : 
that while those whom the merchandiser wishes to reach with his 
message may, and probably do, pass and repass his show-window on 
the street, and linger to look again, thus enabling him to correct mis- 
takes in the original dressing and create a favorable second impression, 
in advertising with a motion picture there is no second impression. 
Those to whom he addresses his show-window of the screen look but 
once, and they never pass that way again. The producer will win or 
lose on that one effort, and that is not a childish matter. 


MR. RICHARDSON: Unquestionably what Mr. Greene says contains much 
truth. I have seen a man buy a 16-mm. projector without knowing anything at 
all about projecting motion pictures; but he got out his instruction book and 

346 C. L. GREENE 

learned how to thread the machine almost correctly and proceeded to give a show. 

MR. OFFENHAUSER: The same situation obtains in the 16-mm. field as in the 
35-mm. field, as far as projection is concerned. We all know that faults exist 
and we all know that when they are made evident those who commit them are 
the ones most likely to correct them. Otherwise, such people do not continue to 
project pictures very long. 

MR. FARNHAM : One point that could well be brought up is that of using lamps 
of the proper voltage. When it is realized that the output of a lamp changes 20 
per cent with a change of only 5 volts, we can readily see how the projection may 
suffer. The professional projectionist uses meters and sees that the current is 
normal; but here again, as Mr. Greene says, it is "only 16-mm.," and "Let the 
voltage go where it will" seems to be the attitude. 

MR. HOVER: The manufacturer and his salesmen furnish the users of equip- 
ment with bulletins describing the advantages of 16-mm. projection, but gives 
them no information whatever concerning the limitations of 16-mm. projection. 
Within my jurisdiction is an auditorium of 1800 seats in which 16-mm. projection 
equipment is installed. The screen is 14 feet wide, the projection distance 120 
feet, and the lamp a 750-watt size. It is easy to imagine the kind of picture and 
the nature of the sound. They are using a 3-watt amplifier. 

MR. TANNEY: The laboratory should not be overlooked. We have had several 
35-mm. negatives reduced to 16-mm., and the results have been very uncertain. 
In some prints the sound is extremely fuzzy; in other prints the pictures jump 
considerably. How that can be remedied, I do not know, but if the laboratories 
would do something about it that would help the situation. 

MR. TOWNSLEY: It is possible to make very good 16-mm. reductions from 
35-mm. negatives, but a tremendous amount of care, skill, and work is 
required. There are laboratories in this country that are doing very fine jobs. 


Summary. Standards have established the exact location of the sound-track of 
motion picture film in relation to the perforations and the picture area. Numerous 
cases are described of prints wherein the sound-track location departs considerably 
from the standard. This is a source of annoyance to the projectionist and often re- 
sults in faults on the screen or in reproduction. 

Examining the standards set up by the Society of Motion Picture 
Engineers, J one is impressed by the very exact placement of the sound- 
track with respect to the sprocket-holes and the picture area. Pro- 
jectionists who must project the combined images of visual and sound 
effects contained in positive prints are well aware of the importance 
of these dimensional standards and of their strict observance in the 
production of prints for use in theaters. 

The dimensions are expressed in thousandths of an inch, and any mis- 
placement of the sound-track with relation to sprocket-holes is ob- 
jectionable from the projectionist's point of view. The projectionist 
adjusts the sound-head elements that guide the film past the sound- 
head aperture in accordance with these dimensions; and unless they 
be adhered to with accuracy by the studio men, the close adjustment 
required will be in error exactly in the amount by which the studio 
misplacement is in error. If the error be sufficient, either the sound 
or the visual effect or both must and will suffer. Many projectionists 
take pride in the effects they present to their audiences, and they 
obviously can not produce perfect effects unless provided with films 
that are of themselves perfect. 

Observations made during the past eight months have shown that 
for some reason there is serious deviation in many prints from the 
standard measurements. What the deviations are and how great 
they are are shown in the accompanying illustrations, which represent 
only a few of the many photographs taken in the projection room. 

* Presented at the Fall, 1937, Meeting at New York N. Y.; received Septem- 
ber 24, 1937. 

** Forum Theatre, Akron, Ohio. 




[J. S. M. P. E. 

The photos are from prints of one large producer. The deviations 
from standard are sufficient to be quite noticeable in the reproduced 
sound in some instances ; in others, very serious harm is done to the 
visual effect upon the screen. 



.30 flM 

FIG. 1. Standard track position. 

Fig. 1 illustrates the track position, which is the same for either 
variable-density or variable-width sound recording. 

Fig. 2 is an enlargement of a section of sound-track of a theater re- 
lease print. Whereas the standard demands that the black line at 
the left impinge upon the edges of the sprocket-holes, in this print the 


line as a whole, and a part of the sound record as well, overlaps the 
sprocket-holes. It is, of course, evident that this has the effect of 
shifting the entire track toward the sprocket-holes, leaving a broad, 
white line between it and the black line dividing it from the picture 

The result of this is the pick-up of some surface noise, which is 
especially objectionable when the sound level is low or when a high 
fader setting is required. If bad enough it may induce motorboating. 

FIG. 2. Enlargement showing sound-track encroach- 
ing upon sprocket-holes. 

FIG. 3. Condition of Fig. 2 aggravated. 

In Fig. 3 the condition shown in Fig. 2 is aggravated, and shows the 
frame lines jutting into the white space left between the sound-track 
and picture area. This causes a very unpleasant background "buzz" 
or hum; in fact, it was necessary with this film to displace the sound- 
head guide-rollers in order to cause the lines to miss the reproducing 
beam. Instead of an 8-mil opaque boundary line between the sound- 
track and the picture area, Fig. 4 shows a transparent space. Many 
times when shifting from one scene to another during projection this 

350 I. GORDON [j. s. M. p. E. 

line also shifts far enough to show upon the edge of the screen as a 
broad strip of white light, appearing and disappearing as the scenes 
change. In one case this occurred three times during one reel. It 
was extremely annoying and highly detrimental to the visual effect. 
Had the line been black, the slight shift of the scene as a whole would 
have passed with but little notice. 

FIG. 4. Enlargement showing transparent space 
between sound-track and picture area. 

FIG. 5. Enlargement showing no separation between 
sound-track and picture area. 

Fig. 5 illustrates the opposite extreme, where the edge of the sound- 
track appears upon the screen, due to the absence of any dividing 
space whatsoever. 

Projectionists are asked to place before the public a sound motion 
picture which is all this great industry has for sale, and to do so in 
the best possible manner. Unlike studio men projectionists are not 
all working in a single plant, and so can not voice their protests in a 
single body. They work either alone or at most in groups of two or 
three. They wish to present the picture and the sound in the best 


possible manner, and real motion picture-sound projectionists take 
great pride in the excellence of the effects, both visual and acoustical, 
that they can place before their audiences. However, excellence can 
not be attained unless the things requisite to attaining it are pro- 
vided. Such inexcusable faults as have been shown tend to lower the 
standards of projection, discourage projectionists from striving for 
high excellence in displaying the salable wares of this great industry. 
It is recommended, therefore, that the Society of Motion Picture 
Engineers take such steps as it may to stop producers from permit- 
ting the release of films containing such faults as have been shown here. 


1 "Dimensional Standards for Motion Picture Apparatus, and Recommended 
Practice," /. Soc. Mot. Pict. Eng., XXIII (Nov., 1934), p. 247. (See Revision of 
Standards, this issue, p. 267.) 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in those magazines that are available may be obtained from the Bibliofilm Service, 
Department of Agriculture, Washington, D. C. 


17 (Dec., 1937), No. 12 

Instantaneous Recording Needles (p. 16). R. H. RANGER 

The Harmonic Producer (p. 19). C. H. BIDWELL 


13 (Dec. 11, 1937), No. 16/17 

Ernophon II mit Schnellanlauf (Ernophon II, with 
Rapid Starter) (p. 193). 

International Photographer 

9 (Dec., 1937), No. 11 
New Track Standards ^p. 24). 

International Photographer 

19 (Jan., 1938), No. 1 

Agfa Introduces Two Super-Fast Motion Picture Nega- 
tive Films (p. 10). W. STULL 

Make-Up Specialist Can Do Much To Assist the Cinema- 

tographer (p. 13). P. WESTMORE 

Engineer Discusses Requirements of True Stereoscopy 
in Motion Pictures (p. 14). G. W. WHEELRIGHT 

Art Reeves Shows New Ultraviolet Recorder (p. 16). 

Bell & Howell Has Novel 8-Mm. Film Viewer (p. 29). 

Bell & Howell Producing Four Film-on-Sound Projec- 
tors (p. 31). 

International Projectionist 

12 (Dec., 1937), No. 12 
New Servicing Tools Needed for Visual Projection 

Equipment (p. 7). A. C. SCHROEDER 

Television Problems A Description for Laymen (p. 11). A. VAN DYCK 
Analyses of Modern Theater Sound Reproducing Units 

(p. 16). A. NADELL 

Some General Principles of Projection Optics (p. 20). R. H. CRICKS 

Decision Rendered on Battle of Split Seconds Anent 

Frame Aperture Rest (p. 22). 



The Elements of Vision The Basis of Projection (p. 23). W. C. KALB 


19 (Nov., 1937), No. 12 

Aufgaben der Bildforschung fur Film und Ferrjsehen 
(Problems in the Investigation of Subjects for Films 
and for Television) (p. 287). R. THUN 

Stand der Tonfilmtechnik (Present Position of Sound 

Film Technic) (p. 290). H. WARNCKE 

Motion Picture Herald (Better Theaters Section) 

129 (Dec. 11, 1937), No. 11 
Advantages of the New Methods of Servicing Sound 

Equipment (p. 29). A. NADELL 

Violations of Standard Spacing in Release Prints (p. 36). F. H. RICHARDSON 

Photographische Industrie 

35 (Dec. 8, 1937), No. 49 

Hochsdrucklampe als photographische Lichtquelle 
(High Pressure Lamps as a Photographic Light 
Source) (p. 1329). P. HATSCHEK 

35 (Dec. 15, 1937), No. 50 

Der neue Raumfilm, System Zeiss Ikon (New Stereo- 
scopic Film, Zeiss Ikon System) (p. 1353). 


10 (Dec., 1937), No. 118 

An Electronic Light Relay for Large Pictures (p. 716). 
British Television through American Eyes (p. 721). 





Officers and Committees in Charge 

W. C. KUNZMANN, Convention Vice-President 

J. I. CRABTREE, Editorial Vice-President 

G. E. MATTHEWS, Chairman, Papers Committee 

W. WHITMORE, Chairman, Publicity Committee 

E. R. GEIB, Chairman, Membership Committee 

N. D. GOLDEN, Chairman, Local Arrangements Committee 

Local Arrangements and Reception Committee 


N. D. GOLDEN, Chairman 





Registration and Information 

W. C. KUNZMANN, Chairman 



Ladies' Reception Committee 

MRS. R. EVANS, Hostess 

assisted by 




Banquet Committee 

R. EVANS, Chairman 



Publicity Committee 

W. WHITMORB, Chairman 




Convention Projection Committee 

H. GRIFFIN, Chairman 



Officers and Members of Washington Projectionist Local 224 

Apparatus Exhibit 

S. HARRIS, Chairman 

Membership Committee 

E. R. GEIB, Chairman 


Hotel and Transportation Committee 

J. G. BRADLEY, Chairman 




The headquarters of the Convention will be the Wardman Park Hotel, where 
excellent accommodations are assured. A reception suite will be provided for the 
ladies, for whom also is to be arranged an interesting program of entertainment. 

By special arrangement with the Hotel Management, special breakfast, lunch- 
eon, and dinner service will be provided on the Continental Room Terrace, 
for SMPE delegates only. 

The following daily hotel rates, European plan, are guaranteed to SMPE 
delegates attending the Convention: 

One person, room and bath $ 3 . 50 

Two persons, standard bed 5 . 00 

Two persons, twin beds 6 . 00 

Parlor suite, one person 9 . 00 

Parlor suite, two persons 11 . 00 

356 SPRING CONVENTION [j. s. M. P. E. 

Room reservation cards will be mailed to the membership of the Society in the 
near future, and those who plan to attend the Spring Convention should return 
their cards promptly to the Wardman Park Hotel to be assured satisfactory 
accommodations. Local railroad ticket agents should be consulted with regard to 
trains and rates. 

For those who will motor to the Convention ample free parking space is avail- 
able on the Hotel grounds. For those who prefer parking in the Hotel garage, 
a special rate of 75 cents a day has been arranged. 

Technical Sessions 

An attractive and interesting program of technical papers is being assembled 
by the Papers Committee. All technical sessions, apparatus symposiums, and 
film programs will be held in the Little Theatre of the Hotel. 

Apparatus Exhibit 

An exhibit of newly developed motion picture apparatus will be held, to which 
all manufacturers of equipment are invited to contribute. No charge will be 
made for space. Information concerning the exhibit and reservations for space 
should be made by writing to the General Office of the Society. 

Apparatus displayed should be newly designed or developed, or should have 
features of technical interest for the engineers attending the Convention. 

Registration and Information 

The Convention registration headquarters will be located at the entrance of the 
Little Theatre, where all the technical sessions will be held. The members of the 
Society and guests attending the Convention are expected to register and receive 
their badges and identification cards for admittance to special evening sessions. 
These cards will also be honored at several de luxe motion picture theaters in 
Washington during the four days of the Convention. 

Informal Luncheon and Semi-Annual Banquet 

The usual informal Luncheon will be held at noon of the opening day of the 
Convention, April 25th, in the Continental Room of the Hotel. On the evening 
of Wednesday, April 27th, will be held the Semi-Annual Banquet of the Society, 
also in the Continental Room, at 8:00 P.M. Addresses will be delivered by 
prominent members of the industry, followed by dancing and other entertain- 

Mar., 1938] SPRING CONVENTION 357 

Motion Pictures 

Delegates registering at the Convention will be supplied with complimentary 
passes to the following motion picture theaters in Washington during the dates of 
the Convention: 

By courtesy of Mr. J. J. Payette: Warners' Uptown and Earle Theaters. 

By courtesy of Mr. H. Meiken: RKO Keith's Theater. 

By courtesy of Mr. C. Barren: Loew's Capitol, Palace, and Columbia Theaters. 

Ladies' Committee 

A number of interesting events are being planned by Mrs. R. Evans, Hostess, 
and the Ladies' Committee. On Monday, April 25th, at 5 P.M. Mrs. Franklin 
D. Roosevelt has kindly consented to receive the ladies of the Convention at the 
White House. All those who intend to be present at the reception should trans- 
mit their names as early as possible to Mr. W. C. Kunzmann, Convention Vice- 
President, at the General Office of the Society, Hotel Pennsylvania, New York, 
N. Y. 


The Wardman Park Hotel management is arranging for golfing privileges for 
SMPE delegates at several courses in the neighborhood. Regulation tennis 
courts are located upon the Hotel property, and riding stables are within a short 
distance of the Hotel. Trips may be arranged to the many points of interest in 
and about Washington. 

Points of Interest 

To list all the points of interest in and about Washington would require too 
much space, but among them may be mentioned the various governmental 
buildings, such as the Capitol, the White House, Library of Congress, Depart- 
ment of Commerce, U. S. Treasury, U. S. Bureau of Standards, Department of 
Justice, Archives Building; and other institutions such as the National Academy 
of Sciences, the Smithsonian Institution, George Washington University, Wash- 
ington Cathedral, Georgetown University, etc. In addition may be included the 
Lincoln Memorial, the Washington Monument, Rock Creek Park, The Francis 
Scott Key Memorial Bridge, Arlington Memorial Bridge, the Potomac River, 
and Tidal Basin. Mt. Vernon, birthplace of Washington, is but a short distance 
away and many other side trips may be made conveniently via the many highways 
radiating from Washington. 



At a meeting of the Section held on February 9th at the meeting rooms of the 
Western Society of Engineers, Chicago, 111., Mr. H. A. DeVry of the DeVry 
Corporation, Chicago, presented a paper on "A New Mechanical Movement 
Developed for a Framing Device for the 35-Mm. Projectors." The meeting 
was well attended and an interesting discussion followed the presentation. 

The next meeting of the Section is scheduled for March 15, 1938. 


The regular monthly meeting of the Section, held on February 23rd, took the 
form of a visit and tour through the laboratories of Consolidated Film Industries, 
Inc., at Fort Lee, N. J. The members were conducted through the plant in 
groups under the guidance of a member of the laboratories, who explained the 
various features of the equipment and processes. The meeting was very well 
attended and considerable interest was indicated by the number of questions 
put to the guides by the members. 


At a meeting held at the Paramount Building, New York, N. Y., on January 
20th, much consideration was given to the condition of release prints received 
at theaters, and a sub-committee was appointed for the purpose of looking into 
the questions of lack of transparency of numbers on Standard Release Print 
Leaders, which makes it difficult to thread film in frame without a framing light ; 
the occurrence of change-over marks at beginnings of fade-outs instead of at the 
middle, or of numbering the footage on the leaders from the point where the 
picture appears after fade-in rather than from the start of the fade-in; out-of- 
focus prints; and feature films running less than sixteen minutes each. 

Other discussion revolved about difficulties with edge-waxing; condition of 
2000-ft. metal reels; the use of a noise-level meter for monitoring sound in 
theaters; and of correct film tension during projection. 

Interim reports were presented by the Sub-Committees on Theater Structures, 
and Projector Output and Screen Illumination, and the following definitions were 
approved by the Projection Practice Committee: 

"A projectionist is a qualified person professionally in operating charge of motion 
picture projection equipment." 

"A projection room is a suitably arranged and protected enclosure wherein is 
placed motion picture projection equipment for professional use." 

The next meeting of the Committee is scheduled for February 17, 1938. 




At a recent meeting of the Admissions Committee, at the General Office of the 
Society, the following applicants for membership were admitted to the Associate 


Cinesound Productions Pty., Ltd., 

Bondi Junction, N. S. W. 
2027 Bleury St., 

Montreal, Canada. 
The Knoll, 

Waldoboro, Maine. 

105 Maryland Ave., 

Palmyra, N. J. 
2412 Cortelyou Rd., 

Brooklyn, N. Y. 

4518 N. Clarendon Ave., 

Chicago, 111. 
Dufaycolor, Inc., 
30 Rockefeller Plaza, 

New York, N. Y. 

Philips Glowlamps Works, 

Eindhoven, Holland. 
8120 Merrill Ave., 

Chicago, 111. 
HOLT, L. J. 

5029 Tennessee Ave., 

St. Louis, Mo. 
Burggasse, 67, 

Vienna VII, Austria. 

RCA Manufacturing Co., 

Camden, N. J. 
KATZ, L. J. 

905 Clark Building, 
Pittsburgh, Penna. 


311 W. 44th St., 

New York, N. Y. 
862 E. 169th St., 

The Bronx, N. Y. 
LONG, J. R. 

1514 McCormick Ave., 
Washington, Indiana. 
MOLE, J. E. 

3847 Randolph Ave., 

Oakland, Calif. 

Acoustical Eng. Co., 
8461 Melrose Ave., 

Los Angeles, Calif. 
2060 Fargo Ave., 

Chicago, 111. 

Av. Neuvo Leon 68, 

D. F., Mexico. 
4913 N. 13th St., 

Philadelphia, Penna. 
67 Barrington Ave., 
Hurstville, N. S. W. 


4846 Rosewood Ave., 

Hollywood, Calif. 

National Carbon Proprietary, Ltd. 
P. O. Box 24, Mascot, Sydney, 


1393 Lexington Ave., 

New York, N. Y. 

424 Patterson Building, 
Denver, Colo. 




6719 Templeton St., 

Huntington Park, Calif. 
55 Rue Vifquin, 
Brussels, Belgium. 

WARE, H. R. 
844 6th Street, 

Portsmouth, Ohio. 
37 Evergreen Lane, 
Haddonfield, N. J. 

In addition, the following applicants have been admitted by vote of the Board 
of Governors to the Fellow and Active grades: 

2922 Davenport Ave., 

Davenport, Iowa. 
BEAL, R. R. (F) 

Radio Corporation of America, 
30 Rockefeller Plaza, 

New York, N. Y. 
CARTER, C. C. (M) 
Filmcraft Laboratories, 
35 Missenden Road, 

Camperdown, N. S. W., 


DEMMER, A. H. (M) 
111 Chestnut St., 
Audubon, N. J. 

Farnsworth Television Inc. of Pa., 
127 E. Mermaid Lane, 
Philadelphia, Penna. 

HANSON, O. B. (M) 
184 Compo Rd., 
Westport, Conn. 

IVES, C. E. (M) 

Eastman Kodak Company, 
Rochester, N. Y. 


Haddonfield Manor, 
Haddonfield, N. J. 

5512 Mullen Ave., 
Los Angeles, Calif. 

SHARP, E. R. (M) 
2302 Scott St., 
Davenport, Iowa. 




Volume XXX APRIL, 1938 Number 4 


A Modern Motion Picture Laboratory C. L. LOOTENS 363 

Light Control in Photography G. MILI 388 

Spectral Distributions and Color-Temperatures of the Radiant 
Energy from Carbon Arcs Used in the Motion Picture In- 
dustry F. T. BOWDITCH AND A. C. DowNES 400 

Newer Types of Stainless Steel and Their Application to Photo- 
graphic Processing Equipment H. A. SMITH 410 

Air-Conditioning with Lithium Chloride G. A. KELLEY 422 

Die-Castings for Photographic Appliances J. C. Fox 432 

The Activated Alumina System as Applied to Air-Conditioning 
and Drying Problems G. L. SIMPSON 449 

New Motion Picture Apparatus 

The Sound-Level Meter in the Motion Picture Industry 


Complete Cue-Mark Elimination and Automatic Change- 
Over S. A. MACLEOD 463 

Current Literature 467 

Washington Convention, April 25th-28th, Inclusive 

General 470 

Abstracts of Papers and Presentation 474 

Society Announcements 498 





Board of Editors 
J. I. CRABTREE, Chairman 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West-Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1938, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 


*President: S. K. WOLF, RKO Building, Rockefeller Center, New York, N. Y. 
*Past-President: H. G. TASKER, Universal City, Calif. 
*Executive V ice-President: K. F. MORGAN, 6601 Romaine St., Los Angeles, 


^^Engineering Vice-President: L. A. JONES, Kodak Park, Rochester, N. Y. 
* Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 
"Financial Vice-President: E. A. WILLIFORD, 30 E. 42nd St., New York, N. Y. 
4 'Convention Vice-P resident: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 
*Secretary: J. FRANK, JR., 90 Gold St., New York, N. Y. 
^Treasurer: L. W. DAVEE, 250 W. 57th St., New York, N. Y. 


*J. O. AALBERG, 157 S. Martel St., Los Angeles, Calif. 

*M. C. BATSEL, Front and Market Sts., Camden, N. J. 

**R. E. FARNHAM, Nela Park, Cleveland, Ohio. 

*G. FRIEDL, JR., 90 Gold St., New York N. Y. 

*A. N. GOLDSMITH, 444 Madison Ave., New York N. Y. 

**H. GRIFFIN, 90 Gold St., New York, N. Y. 

**A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

*S. A. LUKES, 6145 Glenwood Ave., Chicago, 111. 

Term expires December 31, 1938. 

**Term expires December 31, 1939. 



Summary. A description of the new laboratory of Consolidated Film Industries , 
Inc., at Hollywood, Calif., completed during the winter of 1936-37. Included are 
lay-outs and illustrations of equipment on several of the floors. The description oj the 
laboratory and equipment follows the sequence of negative development, dailies, 
master and release printing, together with a description of the special printers, proc- 
essing units, chemical system, silver-recovery system, and other mechanical items 
of interest. 

The lay-out of the various buildings constituting the new Consoli- 
dated Film Laboratory at Hollywood, Calif., is shown in Fig. 1. A 
photograph of the Seward Street elevation of the laboratory and serv- 
ice building is shown in Fig. 2. The laboratory proper is constructed 
of reinforced concrete, and all laboratory workrooms in which film is 
handled are provided with ceiling explosion vents, which in case of 
fire automatically open for the escape of gases. In the daylight work- 
rooms, the vents are provided with frosted glass for skylight illumina- 
tion. In the darkrooms, the glass is made opaque. In case of ex- 
plosion the glass will break, allowing the gases to escape. In order to 
minimize the problems of air-conditioning and cleanliness, no win- 
dows are provided in the laboratory proper, all daylight being ob- 
tained through the glass in the explosion vents. Emergency exits 
are placed at strategic locations in all workrooms. All rooms in 
which film is handled are completely air-conditioned by means of an 
adequate refrigerating and air-conditioning system. 

In laying out the developing machines it was felt that a machine 
that could be used for either negative or positive development would 
conserve floor space and render greater flexibility in the kinds of 
work each machine could produce. The developing machines were 
therefore provided with individual negative and positive developing 
sections. The same hypo and water sections and drying cabinets 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received October 
8, 1937. 

** Republic Productions, Inc., North Hollywood, Calif. 




[J. S. M. P. E. 

were used for both types of film. Since the completion of the labora- 
tory these arrangements have worked out very satisfactorily, due to 
the fact that on peak loads of positive or negative all machines can be 
used for the kind of work required. 

The workrooms necessary for preparing the film during printing 
and developing were arranged around the developing room, so that 
the film could travel from room to room in accordance with the nor- 

FIG. 1. Plot plan of laboratories of Consolidated Film Industries, Hollywood. 

mal sequence of operations. This reduced the floor space required 
by the laboratory. All mechanical equipment was placed in a full- 
sized basement, an arrangement which resulted in a minimum of pip- 
ing and air-conditioning ducts. 

Projection rooms, tinting and toning machines, and miscellaneous 
offices are grouped on the first floor of the service building. Cutting 
rooms, title and optical department, and a third projection room are 
located on the second floor. Detailed lay-outs of all workrooms, to- 
gether with the major equipment, are shown in the floor plans, Figs. 
3, 4, and 5. 


By referring to Fig. 3 it will be seen that the laboratory was laid 
out to minimize the handling of film between operations. Positive 
raw stock is received at the loading platform shown on the upper left- 
hand side of Fig. 3. The film is unboxed and passed into a positive 
storage vault, where it is stacked according to the types of film in stand- 
ard vault racks. Adjacent to this vault is a negative mounting 
room, which contains steel tables with linoleum tops, on which are 
mounted specially designed flange rewinds. 

The printing room is equipped with standard Bell & Howell printers 
which are adjustable for sound-track, standard picture, and full aper- 
ture printing ; Cinex testing machines for making positive tests from 
negative scenes; and duplex color printing machines for printing bi- 

FIG. 2. Exterior view. 

pack negative on double-emulsion positive film. A Duplex printer 
equipped with pilot pins is provided for registration and process 
prints. For conveniently handling the negative and positive film 
cans, individual racks are located at each printer. 

Adjacent to the printing room is a sound-track mounting room 
equipped with steel linoleum-topped tables. In this room the nega- 
tive sound-track is inspected for mechanical defects, and track tests 
are pulled and mounted for developing. An Eastman IIB sensitom- 
eter is also located in this room for making all sensitometric expo- 
sures. The corridor adjacent to these rooms is provided with nega- 
tive and positive safelights. The main corridor at this end of the 
building is dimly lighted so that there is no danger of fogging unde- 
veloped film. 

Six developing machines are installed in groups of three, each group 



[J. S. M. P. E. 

being located in a separate room. Both developing rooms are equipped 
with negative and positive safelights. The outside developing 
room is primarily used for negative action development, while the in- 
side room is used for negative sound-track development. Inasmuch 
as the continuous printers are located in the inside room, practically 
all dailies and master prints are processed in them. The machines in 
the outside room are also used for developing positives printed in the 
printing room. In case of peak loads it is possible to develop action 

FIG. 3. First-floor plan. 

negative or sound-track negative in all six machines. It is, of course, 
the normal procedure to use all six machines for positive development. 
A passageway between the developing room and the drying room 
serves as a light-lock, a feature making it possible to operate the dry- 
ing cabinets in white light. 

A negative cleaning room and two rooms each equipped with an in- 
spection drum are located adjacent to the main corridor. The nega- 
tive cleaning room is provided with steel linoleum -topped tables with 
rewinds mounted on the tables so that the film may be wound by hand 
and drawn through a cloth saturated with a high-test cleaning fluid. 
A vacuum pipe immediately back of the film draws away the fumes 



of the cleaning fluid and assists in the evaporation of the fluid from 
the film. The inspection rooms are equipped with large inspection 
drums 4 x /2 feet in diameter and 9 feet long. Each inspection drum 
is driven by an electric motor having an electromagnetic brake that 
will stop the drum within an angle of about 30 degrees. A foot treadle 
extending the entire length of the drum and within easy reach of 
the operator is provided so that the drum may be started or stopped 
at will. Adequate lights are placed above the drum so that it is pos- 
sible for inspectors to detect the slightest abrasion of the film. All 
negative film is wound on these drums and inspected for mechanical 



Laboratory Hoof 

bplorton vents 

FIG. 4. Second-floor plan. 

defects, a procedure that is followed for all negative entering and leav- 
ing the laboratory. 

Adjacent to these rooms is a large negative assembly room (Fig. 6), 
at one end of which are located large steel linoleum-topped tables 
where newly developed sound-track and action negative is inspected. 
At the same time the "no print" or B negative is removed from the 
negative rolls, after which the "print" or A negative is passed to the 
splicing tables at the other end of the room, where the various nega- 
tive scene rolls are assembled into 1000-ft. rolls in preparation for 
printing dailies or rushes. 

The splicing tables are of desk design and are made of steel with 
linoleum tops. The splicers are of the standard Bell & Howell type, 



[J. S. M. P. E. 

provided with specially designed lighting fixtures for proper inspec- 
tion and assembly of the negative. Coding machines for foot-num- 
bering assembled negatives are also provided. One coding machine 
is designed to number both edges of the negative in one passage 
through the machine. 

A negative timing room is adjacent to the negative assembly room. 
This room is equipped with two light-boxes approximately sixteen 
feet long. Negative test-strips and Cinex positive test-strips can be 
laid on these light-boxes and inspected and timed. 

Next to the timing room is located the control room, which is pro- 

3 aamajll M* 

FIG. 5. Basement plan. 

vided with densitometers, microscopes, and a wall panel of recording 
and indicating instruments. From the readings obtained in this room 
the functioning of machines and processes in the laboratory can be 
determined. Fig. 7 shows the large meter panel on which all record- 
ing and indicating instruments are mounted. In the top row are four 
Leeds & Northrup indicating temperature recorders for the develop- 
ing solutions and a temperature recorder for hypo. The developing 
solution instruments indicate and record within 0.05 degree Fahren- 
heit. In the lower row from left to right are : a Leeds & Northrup 
12-point recorder for wet- and dry-bulb temperatures of the six drying 
cabinets ; a recorder for temperature of wash water ; an indicating gauge 
for air-pressure; a recording voltmeter for the printing lamp genera- 



tor; an indicating gauge for steam pressure; a recorder for tempera- 
ture of ice-water used for air-conditioning and solution cooling; and a 
Leeds & Northrup 6-point recorder for the film speed of the develop- 
ing machines. 

By again referring to Fig. 3, it will be noted that six projection in- 
spection rooms are located at the end of the drying cabinets. In these 
rooms are located continuous projectors provided for projecting 
the positive film as it leaves the drying cabinets. By means of con- 
veyors the inspected film is passed into an automatic stacking ma- 
chine located in the positive finishing room. This room is equipped 

FIG. 6. Negative assembly room. 

with table-type splicers similar to those in the negative assembly 
room. A storage vault is located next to the finishing room, for stor- 
ing film between operations. A shipping room is adjacent to the fin- 
ishing room. 

Two preview rooms are located near the shipping room. One of 
these rooms is equipped with RCA projectors and sound equipment 
(Fig. 8), while the other room is equipped with Powers projectors and 
Western Electric sound equipment. The former room has a large 
screen for reviewing master prints ; the latter has two smaller screens 
so that the two machines may be run simultaneously for the inspec- 
tion of dailies. 

Engineering and laboratory superintendent's offices and rooms for 
customers' men are located on the first floor (Fig. 3). Adjacent to 

370 C. L. LOOTENS [j. S. M. p. E. 

preview room No. 2 is a film-treating room for removing celluloid 
scratches from negative or positive film. A machine specially de- 
signed and patented to accomplish this on action negative that has 
been scratched on the celluloid side, provides effective resurfacing so 
that the scratches will not show when either a contact or an optical 
print is made. 

The susceptibility of green film to scratches and abrasions is common 
knowledge. A protective treatment for toughening and lubricating 
the emulsion, enabling it to resist scratching as well as or even better 
than thoroughly seasoned film, has been developed in the research 

FIG. 7. Control room. 

laboratories of the company at Fort Lee, N. J. In the case of positive 
prints, it has been found that film treated by this process survives its 
early bookings with appreciably less damage to the emulsion than un- 
treated film. Negative film that has been subjected to the protective 
treatment may be handled, cleaned, and spliced with less chance of 
scratching or abrading the emulsion. The processing fluid is applied 
to the film either in the drying cabinets of the developing machines 
or in a special machine as an after-treatment. Application is by 
means of a cloth-covered roller rotating slowly in a direction opposite 
to the travel of the film. 

A completely equipped plant for manufacturing color-prints (Mag- 
nacolor) is situated in the west wing of the first floor (Fig. 3). Posi- 



live film sensitized on both surfaces is used in this process, the color 
record of each component of a bipack negative being exposed on either 

After ordinary black-and-white development, the double-coated 
print is converted into a color picture by dye-toning the side printed 
from the front element of the bipack with an orange dye and convert- 
ing chemically the other image into a transparent blue image. The 
machines used for the production of color-prints are readily con- 
verted into toning machines for producing the recently revived tinted 
and toned effects such as sepia and blue tones. 

FIG. 8. Projection room No. 1. 

Work is usually routed through the laboratory in the following 
sequence : Undeveloped negative is received and checked in the ship- 
ping room. The film is then transported to the negative mounting 
room, where it is inspected by hand for mechanical defects and tests 
are detached from the negative rolls. The tests are then mounted 
on reels and passed to the developing room. There they are devel- 
oped at standard negative developing time, after which they are taken 
to the timing room where they are inspected by the customers' men, 
who make any changes in negative development time necessary. 
The negative rolls in the mounting room are then assembled ac- 
cording to the corresponding development times and the negatives 
developed The rolls are taken from the drying cabinets of the ma- 

372 C. L. LOOTENS [J. S. M. P. E. 

chines and are moved to the negative breakdown tables, where "print" 
and "no print" scenes are separated. The "no print" scenes are 
placed in cans and delivered to the respective studios. The "print" 
scenes are assembled in numerical sequence in the negative assembly 
room, whence they are sent to the printing room where Cinex test 
prints are made from a section of each scene. The Cinex test prints 
are then developed and taken to the timing room, where they are 
separated and placed on the timing table. The timer then reviews 
the test prints and assigns the proper printing light for each scene. 
While the Cinex test prints are being developed, the negative roll is 
sent to the inspection room where it is placed upon an inspection drum 
and inspected for mechanical imperfections. After inspection, the 
negative is thoroughly cleaned and placed into cans awaiting the tim- 
ing control strips for the printers. When the timing control strips 
are available, the negative rolls are sent to the continuous processing 
machines, where they are printed and the positive developed. The 
positive print is inspected on the continuous projector and the film is 
then sent by conveyor to the finishing room. 

After the positive has been assembled properly, it is then available 
for projection in the review room by customers' representatives. 
After the review inspection the film is sent back to the shipping room. 

All dailies, master, and release prints, with the exception of small 
re-orders and replacements, are handled by the continuous processing 
machines, into which raw stock is fed at one end and from which, 
forty minutes later and without a stop, a perfectly inspected print is 
available at the other end. This process continues at the rate of more 
than 6000 feet per hour per machine. The continuous processing ma- 
chines combine into one operation printing, developing, fixing, wash- 
ing, drying, and projecting finished prints. The machine consists of 
specially designed unloading cabinet, splicer, and loading elevator 
which enable an operator to splice on successive rolls of raw stock, per- 
mitting the raw stock to pass to a printing machine as a continuous 
strip. By means of a specially designed printing machine, the pic- 
ture and sound-track are printed continuously on the positive raw 
stock. The printer is designed to operate at a higher speed than the 
developing machine, so that at the end of each negative reel the print- 
ing machine can be stopped to change negatives. Because the print- 
ing machine runs faster than the developing machine, a continuously 
running film-storage elevator is provided between the printing machine 
and the developing machine. While the printing machine is run- 


ning, the elevator accumulates the excess film between the printer and 
and developing machine. While the printer is stopped for re thread- 
ing negative reels, the developing machine draws film from the con- 
tinuously running storage elevator. By this means, the developing 
machine is provided with a continuously running strip of film, and 
time has been allowed for the printing operator to stop the machine 
for changing negatives. The developing, fixing, washing, and drying 
are accomplished in an improved type of Spoor-Thompson developing 

At the discharge end of the dryer, a continuously running film- 
storage elevator is provided so that film may be passed through it to a 
projector where an operator projects and inspects the finished posi- 
tive. The projector operates at a higher speed than the developing 
machine, so that when the projector is running, the excess film is pro- 
vided by the storage elevator, a condition allowing time for the 
operator to stop the projector and examine the film by hand for me- 
chanical defects. It also provides sufficient time for him to stop at 
the end of each reel of film and break the film at the start of successive 
reels. The reel of film is then placed into a can and sent by conveyor 
to the finishing room. 

The loading stand consists of two 2000-ft. magazines for loading 
raw stock. Each magazine is equipped with a signal bell to indicate 
when the roll of film has reached 200 feet. At the end of each roll of 
raw stock the next roll is spliced on a standard Bell & Ho well splicing 
machine. In order to feed the raw stock continuously to the printer, 
a loading elevator is released, allowing excess film to be taken up on 
the elevator. The capacity of the elevator allows one minute for 
splicing, during which time film is supplied to the printers by the 
elevator. A synchronizing mechanism between the elevator and the 
splicer guarantees that splices are made on frame lines. 

The Consolidated continuous automatic printer (Fig. 9) is of the 
unit design type, each unit complete in itself and each printer requir- 
ing a picture head, non-slip track head, two control-strip units, two 
light-changer units, two light-changer electrical control units, six 
positive and negative cleaners, and four take-up units. All units are 
mounted on a main welded steel plate body frame containing the 
driving mechanism for the various units. 

The printing machine is reversible for forward or backward print- 
ing, so that it becomes unnecessary to rethread the negative reels 
when printing successive prints of the same negative. The picture 



[J. S. M. P. E. 

FIG. 9. Automatic printer. 



head is a Consolidated high-speed continuous printing head, with 
sprockets cut with the utmost precision. A special printing gate as- 
sures perfect contact and no reflection, while a narrow printing aper- 
ture produces increased definition. A high-intensity mercury- vapor 
lamp is used as the printing light-source, with a special filter in the 
picture-head optical system that produces the same contrast on high- 
lights and lowlights. 

The track head is designed on the non-slip principle. The light- 
source used for the picture head is also used for the track head. A 
choice of filters in the track head optical system permits ultraviolet 

FIG. 10. Developing machine. 

as well as white-light printing. Faithful reproduction of negative 
recording is accomplished through non-slip, high-pressure contact 
printing. A separate positive and negative guiding system at the 
track-printing aperture allows regulation of track placement. Elimi- 
nation of printer flicker and sprocket flutter is accomplished by a 
carefully designed mechanically filtered drive and properly sized 
loops of film, between the negative feed and hold-back sprockets and 
printing apertures. All track negative splices are automatically 
blooped with a bloop of controllable length and taper. 

376 C. L. LOOTENS [J. S. M. P. E. 

The automatic light-change operates from a control film-strip. 
By punching the proper combination of holes in the control film-strip, 
the proper aperture can be inserted automatically into the optical 
system, thus controlling the printing-light intensity. Dual negative 
contactors provide a positive check on the operation of the light- 
changer, and automatically stop the printer in case of trouble. A 
complete signal-light system informs the operator of any irregular 
operations of the machine. 

The operation of the machine is entirely automatic. The negative, 
raw stock, and control-strips are threaded by the operator. All film 
is threaded tightly over loop-setting rollers, and when the machine is 
started the loop rollers are automatically set in the operating position 
by a solenoid-and-link system. After the operator threads the ma- 
chine he pushes the starting button and does not again touch the 
machine until it automatically stops at the end of the reel. A one- 
way drive clutch makes it possible for the take-ups to act alternately 
as take-ups and take-offs. Their driving tension is adjustable and 
the flanges used are "floating" so that the energy of their rotation has 
no effect upon the film. This insures a very "soft-acting" take-up. 
Pressure and vacuum cleaners keep the negative and raw stock speck- 
proof. The cleaners are located so that all negative and raw stock 
is vacuum cleaned before passing over the printing apertures in either 

To eliminate all extraneous mechanical oscillations a separate 
motor is used to drive the two printing heads. The take-ups and 
control-strip units are driven by the main motor and transmission. 
The power supplied to the printer is 120- volt d-c. for relay control of 
the automatic light-changer; 500- volt d-c. for the mercury- vapor 
printer lamp; and 220-volt, 3-phase a-c. for the motor drive and 

To allow the operator time to rethread the printer after each print 
and maintain continuous development, a printing elevator is located 
between the printer and the developing machine. The printer op- 
erates at a speed of 180 feet a minute, whereas the speed of the de- 
veloping machine is between 110 and 130 feet a minute. While the 
printer is running the elevator takes up the excess film and discharges 
it to the developer when the printer is stopped. During this time the 
elevator stores up approximately 350 feet of film while the printer is 
running, allowing 2 1 /z to 3 minutes for changing negatives. 

The developing machine (Fig. 10) is an improved Spoor-Thompson 


type. It is constructed of stainless steel throughout, including the 
solution tanks. The machines consist of two tanks for negative de- 
velopment, one tank for positive development, a stop-bath tank, a 
hypo tank, and a wash-water tank with partitions for counterflow 
washing. The wash tank is also equipped with water jets for spray 
washing. Either of two negative developing solutions or two posi- 
tive developing solutions may be used, and the circulation piping is 
such that a solution may be drained out of one tank and replaced by a 
different solution in eight minutes. Rubber squeegees remove ex- 
cess solution from the film as it travels from one tank to the next. 
The machine is driven by a Link-Belt P. I. V. drive with a control for 
changing speeds at the developing end of the machine. A Weston 
generator and tachometer indicates the speed at which the film is 
travelling and records it on a Leeds & Northrup 6-point Micromax re- 
corder on the control room board. The film passes over and rotates 
a rubber roller that is exactly one foot in circumference and is con- 
nected to the Weston generator. A signal system also connected to 
the generator flashes a light and rings a bell, calling the attention of 
the operator to a break in the machine so that it can be fixed without 
loss of film in the developer. A Shepard-Niles electric hoist is at- 
tached by cable to the machine to lift the entire mechanism out of the 
tanks for inspection. At the end of the wash tanks the film passes 
through a partition out of the darkroom into the dry box room. 

The film entering the drybox passes through a vacuum type 
squeegee which removes excess water from the film. The drying 
cabinet drive is part of the developer transmission system. A film- 
waxing and emulsion-processing unit is located in the projector ele- 
vator which is the fourth compartment of the drying cabinet. Here 
the film may be waxed or processed as required, with enough drying 
time in the projector elevator to evaporate the processing fluid. The 
film is projected at 165 feet a minute, and the projector storage eleva- 
tor allows the inspector to stop the projector for examining the film 
by hand and placing the reels into cans. The projector elevator 
(Fig. 11) is also equipped with an improved type of "comealong" and 
film take-up in case it becomes necessary to take up the film without 
projecting. The film, on leaving the drying cabinet, passes through 
two fire-traps on its way to the projector. One trap is located at the 
end of the drying cabinet and one in the projection room wall. In 
case of fire in the projector an electrical contact energizes a solenoid 
in the fire-trap and releases a sharp blade, which cuts the film at the 



U. S. M. P. E. 

projection room wall and at the drying cabinet, thus preventing fire 
from travelling from the projector to the cabinet. The continuous 
projector (Fig. 12) consists of a Simplex double-pin intermittent 
mechanism mounted upon a specially designed table and equipped 
with an individual motor drive. A 400-watt projection lamp is used 
with a reflector and Bausch & Lomb condenser assembly 41-26-13. 
The lamp house is cooled by a 2-inch 3400-rpm. fan. A 2 l / 2 E.F. 

FIG. 11. Drying cabinet. 

lens is used having a throw of 12 feet. A roller conveyor system 
transfers the inspected reels to the finishing room where a vertical 
can-stacker automatically stacks the cans. 

The basement floor plan, Fig. 5, provides space for the chemical 
laboratory, chemical mixing and circulating systems, silver recovery 
system, refrigeration, air-conditioning and ventilating systems, film- 
drying system, temperature and pressure controls, compressors, 
printer generators, water filters and softener, machine shop, boiler 
room, and generator room. 

There are four developing baths : action negative, variable-density 



sound negative, variable-width sound negative, and positive print. 
Each of the four developing baths circulates from a 1000-gallon tank 
through a stainless steel heat interchanger to the developing tanks 
on the first floor. 

By means of an automatic air-operated motor valve, circulation 
rates of 25 gallons a minute for positive and 50 gallons a minute for 

FIG. 12. Continuous projector. 

negative, per machine, are maintained regardless of the number of 
machines open for circulation. This unusually high rate of circula- 
tion makes possible a nicety of temperature control seldom encoun- 
tered in industrial processes. The Taylor Instrument Co. control 
mechanism limits temperature variation to a value or range not 

380 C. L. LOOTENS [J. S. M. p. E. 

exceeding 0.1 F, as indicated and recorded by a Leeds & Northrup 
Micromax recorder in the control room. 

All piping in the developer circulation system is of hard rubber, 
and all parts of a LaBour centrifugal circulating pump that come into 
contact with the solution are of lead or G-60 steel alloy. A spare 
1000-gal. tank is also connected with each system to provide storage 
for new developer which may be mixed in advance. The chemicals 
are dissolved in a 120-gal. Pfaudler glass-lined mixing tank with 
built-in stirring propeller. Two mixing tanks have been provided, 
one for the two low-alkaline developers and the other for the two high- 
alkaline developers. Two high-capacity stainless steel heat inter- 

FIG. 13. Machinery room, basement. 

changers connected with both the mixing tanks rapidly cool newly pre- 
pared solutions. The developing baths are kept free from colloidal 
developer sludge and other suspended matter by continuous filtration 
through Bowser monel metal leaf -type filters using Johns-Man ville 
Filter-eel as the filtering medium. The efficiency with which this 
system removes colloidal silver sludge from the action negative de- 
veloper makes it possible to use a developing formula of very low al- 
kalinity (pH 8.1) producing extremely fine-grain characteristics. 
This type can not be used commercially without filtration because 
of its tendency to sludge rapidly and deposit scum on the film. 

Standard negative action development time is ll l /z minutes at 
67 F. The negative developing solution is a modification of the 
Eastman D-76 formula, containing such concentrations of borax 
and boric acid as to produce a buffered solution at pH 8.1. As dis- 


cussed above the unusually low alkalinity of the developer and the 
relatively long time of development are conducive to fine-grain 
results. Both the test and the time-temperature methods of nega- 
tive development are employed depending upon the instructions of the 
customer. Development at the laboratory's standard time produces 
a gamma of about 0.68 as measured by an Eastman sensitometer 
exposure on Eastman Super-^T film. 

A special developer capable of producing a gamma of about 2.70 
with a minimum of fog is used for variable-width track negative. 
Tests are usually removed from the end of each roll, developed, 
and density measurements made to determine the developing time 
for the complete roll. 

All the variable-density track recordings are developed in one ma- 
chine to an average gamma of 0.38. The developer used is a dilute 
modification of the D-76 formula, with the sulfite concentration re- 
stored to 60 grams per liter. 

Standard positive development time is 4*/2 minutes at 68F in a 
developing bath that is a modification of the Eastman D-16 formula. 
Under these conditions an average gamma of 2.10 is maintained. 

The condition of all developing baths is checked by running sensito- 
metric strips at the standard developing times. The strips are ex- 
posed 12 hours in advance, to reduce to a minimum errors caused by 
loss or growth of the image immediately following exposure. The 
strips are developed every thirty minutes and are read and plotted 
without delay. By this means any tendency of the developing bath 
to deviate from its normal characteristics is detected and changes are 
made in the rate of addition of replenisher before the characteristics 
of the bath have changed appreciably. Development of action 
negative, variable-density sound negative, and positive prints is 
governed by the time-gamma curves obtained from the sensitometric 
strips. All emulsions received by the laboratory are tested sensito- 
metrically and graded according to relative speed and contrast. 
Exposure level and scale of all printing and Cinex testing machines 
are checked daily, using the sensitometer as a constant-exposure 
standard. Exposure levels are adjusted by means of a rheostat and 
voltmeter on each machine. The control of developing and printing 
is therefore effected entirely by sensitometric methods. 

The fixing bath is compounded in accordance with Formula F-5 
of the Eastman Kodak Company, and contains potassium alum as 
the hardening agent. Its fixing power is continuously maintained 



[J. S. M. P. E. 

at the proper level by an automatic hypo and silver-recovery plant, 
and its acidity is likewise continuously maintained by a slow manu- 
ally controlled acetic acid feed at the main circulation tank. The 
time of immersion in the fixing bath is ll l /z minutes. The hypo and 
silver-recovery system installed in the new plant is a modern, com- 
pletely automatic version of the original sulfide precipitation process 
used in the early days of the industry. 

A fully equipped chemical laboratory adjoins the mixing room 
(Fig. 5). Samples of each lot of chemicals received are tested chemi- 
cally and photographically before acceptance. Such tests and ex- 

FIG. 14. Ventilating system and dehumidifier. 

perimental work as are necessary for control of the developing and 
fixing baths are performed by graduate chemists. 

The refrigeration (Fig. 13) for the laboratory is provided by a 100- 
ton Carrier centrifugal refrigeration machine, which cools water in a 
tank located nearby. The tank is part of the building structure and 
has a capacity of approximately 50,000 gallons. The machine is 
equipped with an automatic control for the liquor pump that will 
stop all refrigeration when the desired temperature is attained in the 
tank. There are five pumps on top of the tank : one for cooling and 
maintaining the temperatures of the solutions in the various chemical 
systems; a second for cooling the wash water; a third for air-condi- 
tioning the laboratory; a fourth for conditioning the dryboxes; and 
the fifth for controlling solution temperatures in the color department. 


The air-conditioning system (Fig. 14) for the laboratory proper 
maintains an average temperature of 72 F dry-bulb and 70 per cent 
relative humidity. The distribution system is divided so that the 
two developing rooms will be under one control, the printing and 
negative mounting rooms under a second control, and the negative 
cleaning, negative inspection, negative assembly and control rooms 
under a third control. There is one dew-point control at the air 
washer for regulating the dew-point temperature of the air leaving 
the fan. A thermostat located in the central room of each of the 
three groups listed above controls the volume damper and heater for 
regulating the temperature in each group of rooms. There is a com- 
plete filtering system using airmat pocket-type filters located on the 
discharge side of the fan. 

The color department air-conditioning equipment is located in the 
color room. It is complete with an air washer, fan, filters, heater, and 
controls. A similar condition is maintained here as in the laboratory 
proper. The ventilating system, which supplies the work rooms on 
the first floor that are not conditioned, and the basement, consists of 
a fan, filters, four sets of heaters, and a duct distributing system. 
One heater controls the temperature of the air going into the machine 
shop; a second controls the temperature of the air going into the 
chemical tank and machine room; a third, the temperature of air 
going into the first-floor work rooms; and a fourth, the projection 
rooms. Each heater has a thermostat centrally located in the rooms 
that it heats. The filters of the system are installed in the discharge 
side of the fan unit. A similar system is installed on the second floor 
to heat and ventilate the cutting rooms, title department, and offices. 

Individual exhaust systems serve the negative cleaning room, pro- 
jection booth, six inspection rooms, waxing room, machine shop, 
chemical laboratory, optical printing room, art department, and spray 

The film-drying system consists of an individual recirculating sys- 
tem for each drybox. Each unit consists of a fan with a heater on the 
intake side, mixing casing, and airmat filters. The filters are located 
so that the air is discharged into the drying cabinets immediately 
after passing through the filters. The air enters the cabinet at the 
film-discharge end and, after passing through four compartments in 
the drybox, returns at the film-entering end through the heater to the 
intake of the fan. Each drying cabinet has an automatic temperature 
control. A thermostat located at the end of the cabinet where the air 

384 C. L. LOOTENS [J. S. M. p. E. 

enters, operates a diaphragm valve on the steam line connected to the 
heater. An automatic humidity control which opens a damper to a 
duct of dry air is located at the air-discharge end of the drying cabinet. 
As the air becomes wet due to the moisture removed from the film, 
the hygrostat automatically opens the damper, allowing enough dry 
air to enter the sytem thereby regulating the humidity to the preset 
percentage. The dry air is supplied by a dehumidifier and heater. 
The air is washed and the dew-point becomes approximately the same 
as the temperature of the water in the refrigeration tank, which is 
42 F. By heating this low-dew-point air to about 70 F, a supply of 
dry air (35 per cent humidity) is available. A thermostat controls 
the heater on the intake side of the fan for this common supply for all 
six dryboxes. The film is dried in each drybox at 75-80F and 
45-50 per cent relative humidity. A wet- and dry-bulb, connected 
to the recorder in the control room, is located after the filters in each 
drybox for recording the condition of the entering air. 

Two Chicago pneumatic compressors (Fig. 13), complete with 
coolers, oil and water separators, filters, and storage tanks, supply 
the air for all controls and blowout nozzles. An auxiliary compressor 
supplies the air for the controls when the large compressors are not 

The direct current for the tester and printer lamps is supplied by 
three printer generators. A 6-kw. 120-volt d-c. generator supplies 
the printing room and the d-c. power line to the Consolidated auto- 
matic printers. A 2-kw. 120-volt d-c. generator, connected in paral- 
lel at the switchboard through a disconnect switch, is available for 
emergency use. A 500-volt 3-kw. generator supplies power for the 
high-intensity mercury-vapor lamps used in the automatic printers. 

All water entering the plant for use in washing film or mixing de- 
veloper and hypo is filtered. Three sand-and-gravel filters are lo- 
cated on top of the 50,000-gallon refrigeration tank, which filters 
approximately 150 gallons of water a minute when all developing 
and color machines are operating. A 10,000-gallon Permutit Zeolite 
water softener supplies filtered soft water for mixing developer. 

A modern machine shop is located at the north end of the basement. 
A large tool crib for storing mechanical and electrical maintenance 
supplies and replacement parts is located along the east wall of the 

Adjacent to the machine shop is the boiler room. Two 25-hp. 
Peerless gas-fired boilers supply steam for laboratory heating and 


temperature control. Six pounds of steam is maintained in the boilers 
by automatic control of the gas-supply line. The boilers are equipped 
with automatic feed- water supply devices. 

The generator room consists of a bank of six d-c. generators which 
supply current to the projector arc lamps in the three projection 
booths, and to the optical printers and camera department. A 
McCormick-Deering model Pa-50 power unit directly coupled to a 
22.5 kw. Palmer 3-phase, 60-cycle, 220-volt, electric generator is used 
as an emergency stand-by generator. The power unit is operated 
on natural gas, and is equipped with an electric starter and automatic 
choke. The generator is connected in parallel with the city power 
lines through an automatic electrically operated disconnect switch 
to the continuous process developing machine motors and drying 
cabinet air-circulating fans. In case of power failure on the city 
line, the power unit will automatically start and reach full speed in 
less than 10 seconds and supply power to the developing machines. 
This prevents serious losses that would result with negative in the 
developer should the machines stop when the city power fails. 

The second-floor plan (Fig. 4) shows the location of offices, pro- 
jection room No. 3, cutting rooms, and title department. The cutting 
rooms are for the use of customers, and are fully equipped with mod- 
ern cutting tables, light- wells, bins, racks, rewinds, and moviolas. 
Vault space is provided for the negative and for positive work prints. 
The title department may be subdivided into the art, camera, and 
optical departments. It is in the art department that the ideas for 
main titles originate. The artists blend their creative ability with 
trained hands and furnish the producer with a number of ideas from 
which he chooses his main title. When the producer selects the idea 
for the main title, the artists make the cards and letter the titles and 
screen credits and pass their finished product to the camera depart- 
ment for photographing. This department is equipped for many 
novel effects, and also, in many instances, photographs the production 
inserts and lettering for superimposed titling. An animating stand 
and a dolly for zoom work are included in the equipment. In the 
optical department with its battery of optical printers and their 
operators, are produced the lap dissolves, wipes, superimposed titles, 
montages, transitions, and many other valuable tricks of dramatic 

With the facilities as described, the laboratory can give its cus- 
tomers complete service starting with main title, developing and 

386 C. L. LOOTENS [J. S. M. P. E. 

printing, release printing, inserts, and optical work, and ending with 
end titles. 

The author acknowledges and appreciates the assistance of Messrs. 
S. P. Solow and E. H. Reichard for their assistance in preparing this 


MR. CRABTREE: With regard to silver recovery, how do you filter out the 
precipitated silver sulfide? Is it so colloidal that it does not filter easily, or 
do you so arrange conditions that you get a flocculent precipitate? 

MR. MILLER: We have a 100-gallon tank which is used to allow the particles 
in the solution to coalesce before it goes into the filter press. The filter press is 
fed with one of the Johns-Manville filter aids, to cause a porous cake to form and 
the filtration to continue. We use an ordinary hard-rubber press for the filtra- 

MR. CRABTREE: To what extent are the negatives timed with the so-called 
"timing machines," and to what extent do you still employ the visual method of 

MR. MILLER: There seems to be quite a difference of opinion between the 
East and West Coasts as to the better method of timing. On the West Coast, 
for several reasons, we use the Cinex testing machine to make test-strips before 
the negative is printed. That is required by most of our customers, and the 
cameramen want to see the test-strips. However, here in the East we rely upon 
visual timing inspection of the negatives. That places dependence upon the 
timer's judgment, but it is done with surprising accuracy. 

MR. CRABTREE: This system in which the film goes into the machine at one 
end and comes out ready for the theater at the other end involves a rather long 
train of operations. Suppose there is a breakdown of one machine; then expen- 
sive machinery is tied up until repairs are made. How serious is that factor? 

MR. MILLER: We have not found it serious at all. We have been operating 
continuously at Fort Lee (N. J.) for four years now without any really serious 
breakdown. Of course, breakdown of the developing machine proper is the 
only thing that really ties up the machine completely. We can use the develop- 
ing machine for work from other types of printers that are not connected, in 
case the printing machine fails. We can wind film on a take-up, as you saw, in 
case of projection failure, so we are really limited only by the developing machine. 

MR. NICHOLSON: What is the construction of the rollers in the fixing solu- 
tion? How are the ball bearings mounted, and what is the nature of the roller 

MR. MILLER: The rollers in the bottom of the solution are made of molded 
material, by our own company at Scranton. The bottom roller is a single unit ; 
that is, the entire shaft with rollers is a single unit. There are no ball bearings. 
We use a Celeron bushing for the shaft. The top rollers are merely bushed on a 
stainless steel shaft. They are not fastened to the shaft, or interconnected. 
The bottom rollers, of course, in this kind of machine are molded rollers. 

MR. ROBERTS: Do I understand that the printer runs in either direction? 
Does that mean that half the reels will come off reversed in the projector? 


MR. MILLER: In release printing that is true. First prints from freshly 
developed negatives are always fed through head first. 

MR. M. RICKER: What mechanical patchers are used in patching the film? 

MR. MILLER: We use various types depending upon the situation. We use 
some of the Mercer type of patch, which links in the perforations. 

MR. RICKER: Narrow or wide? 

MR. MILLER: The narrow type, engaging two perforations on each side. 
For emergency breaks in the drying cabinet we also use ordinary paper fasteners. 
The two ends can be linked together very quickly. 

MR. RUTHERFORD: I understand that after precipitation with the sulfide, 
the bath is used again. In my experience, I have found that that is not 
a practical thing to do in small laboratories. 

MR. CRABTREE: As long as no more sulfide is added than is necessary to 
precipitate the silver sulfide, then effectively there is no sodium sulfide in solu- 
tion. The silver sulfide is very insoluble and, therefore, the concentration of 
sulfide ion is extremely low and does not sulfide the silver halide emulsion. If 
you have an excess, then it will fog the film. 

MR. SCHAEFFER: What is the speed of your printing machines? 

MR. MILLER: 180 feet per minute. 

MR. SCHAEFFER: Do you project at 180 feet? 


MR. SCHAEFFER: Is the projection silent? 

MR. MILLER: Silent. Any projection with sound is on separate projectors in 
one of the larger screening rooms. 

MR. RICKER: Do you use the acid fixing bath, or do you harden separately 
from the hypo? 

MR. MILLER: We use a standard formula, Eastman F-5, I believe it is; a 
potassium alum, acetic acid type of fixer. 

MR. BRADFORD: I should like to inquire about the tanks and the other stain- 
less-steel equipment. Are they welded? 

MR. MILLER: The tanks are welded. I am not familiar with the exact 

MR. CRABTREE: What materials are you using for the toning bath tanks? 
Are you using continuous machines for toning, or the rack and tank? 

MR. MILLER: The continuous machine. We use the same machines de- 
signed for the Magnacolor process, in which one side of the film is floated and 
the second toning is carried out by immersion of the double-coated stock. 

MR. CRABTREE : We made some resistivity tests of stainless steel in relation 
to the uranium and the iron baths, and we found that ordinary 18-8 stainless 
steel is surprisingly resistant to both those baths. However, if the iron toning 
bath contains much ferric chloride, the steel goes to pieces very quickly. 



Summary The principles underlying light control by means of various types of 
reflectors and lenses for the attainment of proper light-modeling in photography are 
formulated. The basic units that may be employed are discussed with particular 
reference to their performance, advantages, and limitations. 

The main problem of photographic technic is to reproduce a subject 
with the maximum possible range of contrast without loss of detail. 
Because of the great variety of subjects and the still greater number of 
possible effects, this is best achieved by means of artificial lighting, 
which is constant and can readily be controlled in intensity, direction, 
and spread. While the brightness pattern changes with each set, 
and may appear to be entirely different in practically every type of 
photograph, the lighting of any subject may be divided into three 
main components as illustrated by the three photographs in Fig. 1. 

(a) Front lighting, which represents the minimum intensity required for 
retaining detail in the shadows. 

(&) Side lighting, which provides blunt modeling for a relatively large portion 
of the subject. 

(c) Highlighting, which brings into sharp prominence such small features as 
possess greater significance than their mere size would indicate. 

It is entirely feasible to achieve these three constituents of photo- 
graphic lighting by placing bare lamps at various angles and dis- 
tances. Such an arrangement would, however, be extremely un- 
satisfactory because of rank inefficiency. A bare light-source is 
primarily a converter of electrical energy into light radiated in all 
directions, and, in most cases, only one-tenth or less of its total 
light output is delivered within the angle included in a picture, which 
seldom exceeds 40 to 60 degrees. By employing a reflector or a 
lens, or a combination of both, in conjunction with the light-source, 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received Decem- 
ber 9, 1937. 

** Westinghouse Lamp Division, Westinghouse Electric & Mfg. Co., Bloom- 
field, N. J. 




FIG. 1. 


Components of photographic lighting. 

it is not only possible to project a much larger portion of the total 
light output within the effective photographic angle, but also to 
provide the wide range of beam intensities and angular spreads so 
necessary for interesting modeling. The purpose of this paper is to 







FIG. 2. Types of reflection and transmission, depending 
upon surface finish and structure of the medium. 

390 G. MILI 

set forth the advantages afforded by, and some of the limitations in 
the use of, reflectors and lenses for light control in photography. 

Reflection and Transmission. The manner in which a narrow 
pencil of light striking an extended surface may be redirected by 
reflection is indicated diagrammatically in Fig. 2. If the various 
infinitesimal areas that compose an extended surface are all lined up 
in the same plane, the surface acquires a specular polish, and a pencil 
of light falling upon it is reflected in a given direction. If the struc- 
ture of the surface is such that the infinitesimal areas composing it are 
set at extremely varied angles, the surface acquires a diffuse or matte 
finish, thereby destroying any directional control through reflection. 
There are also surfaces in which the infinitesimal areas are set partly 

(a) Parabolic (b) Spherical (c) Elliptical 

FIG. 3. Geometrical contours for concentrating the 

light radiated by a bare source. 

at various angles and partly in one plane. By this means a certain 
amount of diffusion is achieved without complete loss of directional 
control. This is known as spread reflection, while the surfaces are 
usually designated as semi-matte. 

The manner in which a narrow pencil of light falling upon a glass 
surface may be transmitted is also shown in Fig. 2. It is well to note 
that when diffusion and translucence are achieved by structural or 
surface treatment light is only partly transmitted and partly reflected 
back, thereby reducing the efficiency. 

As pointed out above, the problem in photographic lighting is to 
concentrate light radiated in all directions into a relatively narrow 
beam. This is achieved by employing reflectors having contours 
patterned after geometrical curves, such as the parabola, circle, and 

(a) Plano-convex 

(b) Fresnel 

FIG. 4. The Fresnel construction makes practicable 
short-focus lenses with large pick-up angle without in- 

Lighted reflector Bare light-source 

FIG. 5. A lighted reflector as seen from a 
point in the beam. Note that the increase 
in beam over bare light-source intensity is 
due to an increase in the apparent size of the 

Lo = Distance at which entire reflector area is effective, and represents the 

minimum distance at which inverse-square law may be applied in mea- 

suring maximum beam candle-power. 

Beam Cp = Solid angle = irR* 
Source Cp Solid angle a -n-r 2 


Beam Cp = 


Source brightness X mirror projected area 
FIG. 6. Maximum beam candle-power with parabolic reflector. 

392 G. MILI 

ellipse, as shown in Fig. 3. It is obvious that with matte surfaces 
the contour of the reflector has no important bearing upon its per- 
formance, although poor design may conceivably somewhat reduce 
the efficiency of the combined unit. To a degree this is also true of 
semi-matte reflectors, although here some contours are slightly more 
effective in beam control than others. With specular surfaces the 
reflector contour is the determining factor in beam formation. 

Accurate beam control may also be achieved by means of a clear 
plano-convex lens, which duplicates in every respect the performance 
of a specular parabolic reflector. Such a lens is shown in Fig. 4 (a). 
In order to increase the pick-up angle without increasing the lens 
thickness, a modified form known as the Fresnel type, shown in 
Fig. 4(&), is now more widely employed. With the Fresnel lens it is 
possible, because of the larger pick-up angle, to double and even to 
triple the efficiency usually attained with a plano-convex lens. 

The Parabolic Projector. A parabolic reflector has the well known 
property of reflecting a ray of light, originating at its focal point, 
parallel to its axis. Accordingly, with a point-source placed at the 
focus the beam would be parallel. With light-sources of measurable 
dimensions, however, the beam has a definite angular spread, its 
divergence being equal to the largest angle subtended by the source 
to any point on the reflector surface. 

In Fig. 5 are shown two photographs, one of a bare source and one 
of the source placed at the focal point of a paraboloid. It must be 
realized that the increase in intensity gained by means of the re- 
flector is caused entirely by an increase in the effective source size, 
which now equals the projected area of the reflector. This relation- 
ship is expressed mathematically in the equation embodied in Fig. 6. 
Assuming the reflection factor to be unity, the maximum beam candle- 
power is equal to the product of the light-source brightness and the 
projected area of the reflector. Actually the reflection factor is al- 
ways less than unity, and should be included in the product when 
determining the candle-power attainable with any given reflector. 

With the light-source at or near the focal point, the beam from a 
parabolic reflector is much too narrow to afford sufficient coverage 
except for lighting restricted to very small areas. It is possible, 
however, to increase the beam spread by moving the light-source 
away from the focal position. This movement is relatively slight, 
and does not materially affect the pick-up angle of the reflector or 
the volume of light projected into the beam. Consequently the 

FIG. 7. Diagram of parabolic pro- 
jector with slightly corrugated reflecting 
surface. The useful range in beam 
spread and the change in light-source 
position are indicated. 


16 12 8 4 4 8 12 16 


FIG. 8. Relative intensity curves for parabolic projector 
with slightly corrugated reflecting surface. 

FIG. 9. Deep parabolic reflector 
with semi-matte surface finish, ideally 
suited for front lighting. 



FIG. 10. Relative intensity distribution for 
deep parabolic reflector with semi-matte sur- 
face finish. 



average beam intensity must necessarily drop in proportion to the 
beam spread. It can be established by visual inspection of the re- 
flector from a point along or near the axis of the beam that, as the 
light-source is moved out of focus, the active reflector area that is, 
the area redirecting light to that point decreases, thereby causing 
the corresponding drop in intensity. Furthermore, the increase in 
beam spread is more pronounced at the center than along the edge of 
the reflector, resulting in greater reduction of intensity at the center 

FIG. 11. Diagram of Fresnel lens spotlight, indicating 
useful range in beam spread and change in light-source 

than at the edges of the beam. This is very objectionable, since it 
leads to the formation of a dark area in the center of the beam. The 
simplest method of overcoming this deficiency is to introduce slight 
corrugations on the reflector surface. Such a reflector is shown in 
Fig. 7, while in Fig. 8 are plotted various candle-power distribution 
curves for beam spreads ranging from 14 to 30 degrees. It is obvious 
from the curves that even with corrugations the useful range of this 
paraboloid, which is representative of present practice, is limited to 
beam spreads between 14 and 24 degrees, if even distribution is re- 

For wider spreads with absolute uniformity, deep paraboloids with 
a semi-matte finish are employed. Such units have a fixed lamp 



[J. S. M. P. E. 

position, and the spread may be changed only very slightly by the 
use of different light-sources. A unit of this type is shown diagram - 
matically in Fig. 9, while its performance is given by Fig. 10. 

The Fresnel Lens Spotlight. Greater range of beam spread than is 
usually possible with a corrugated parabolic reflector may be attained 
with the Fresnel type of lens spotlight, a diagram of which is drawn 


24 20 

12 8 4 4 6 12 

20 24 

FIG. 12. Relative intensity curves with Fresnel lens 

in Fig. 11. The relative performance of a representative unit of this 
type is given in Fig. 12, which shows a practically uniform beam dis- 
tribution between 10 and 44 degrees. In order to achieve this, it is 
necessary to make the Fresnel lens with a narrow circular or recti- 
linear divergence by ribbing the plain back surface, a feature that is 
embodied in the unit shown. 

April, 1938] 



FIG. 13. Proper use of spherical 
mirror with monoplane filament. 
(Top) filament and reflector image; 
(bottom) filament alone. 

The Spherical Reflector. A 
spherical reflector is included 
in the diagrams of the two units 
described. As is well known, 
the spherical reflector has the 
property of redirecting a ray of 
light originating at the center 
of curvature back to the same 
point. For this reason the 
spherical reflector is used mostly 
as an auxiliary to redirect stray 
light, which would otherwise be 
lost, back into the main beam. 
The increase in beam efficiency 
provided by a spherical reflector 
will range from 35 to 75 per cent, 
depending upon the reflection 
factor and the type of light- 
source. The proper focusing of a spherical mirror when used in 
conjunction with a monoplane filament lamp is illustrated in Fig. 13. 
The Aperture Spotlight. Occasionally it is necessary to illuminate 
an area well defined in outline without any spill of light. This may 
be achieved by means of a spotlight designed on the principle of a 

motion picture projector. The 
elements of one of the most effi- 
cient units of this type ellipti- 
cal reflector, aperture, and plano- 
convex lens are indicated in 
Fig. 14. Fig. 15 is a photograph 
of a spotlight of this type, lighted 
by a similar unit, showing the 
precise pattern of the lighted 
area in the background. In some cases the elliptical reflector is 
replaced by a shallow paraboloid. A unit equipped with this 
type of reflector gives a lighted area with somewhat sharper 
outlines, but its efficiency is much lower. For versatility, the aper- 
ture is made of vanes which can be set at various angles and moved 
in and out to change both the shape and the size of the lighted area. 
Lighting Considerations. While it is possible by the use of re- 
flectors and lenses to increase the amount of light within the effective 

FIG. 14. 

Optical elements of aper- 
ture spotlight. 

398 G. MILI [J. S. M. P. E. 

photographic angle and to increase intensities to a level permitting 
short exposures, one further factor must be considered. Since the 
lens or the reflector, which becomes the effective light-source, is much 
larger than the bare light-source, the shadows cast by such units are 
of necessity diffuse. Sharp shadows may be attained only by the use 
of bare light-sources, or by placing the source sufficiently far from 
the focal position of the lens or reflector so that only a small portion 
of the reflector or lens is active at any point in the beam. For this 

FIG. 15. Aperture spotlight lighted by 
means of a similar unit. Note the definite 
outline of the projected beam on the background. 

reason, lens spotlights with the source in the "flood" position give 
the sharpest shadows, next to the bare light-sources themselves. 

Theoretically it would be entirely feasible to obtain most of the 
lighting effects required under any given set of conditions by employ- 
ing several units of one kind and merely adjusting their distance from 
the subject. However, working conditions in the studio require two 
or more types of unit, depending upon the size of the set, the dis- 
tance at which the units may be placed, and the effect to be achieved. 
Because of their narrow beam spread and extreme intensity, parabolic 
projectors are used to the greatest advantage for long throws and 
strong highlighting. The lens type of spotlight comes into its own 


in highlighting at close range and in floodlighting large areas from 
medium distances, although with the present trend toward large 
heat-resisting lenses it may supersede the parabolic reflector for most 
applications. The semi-matte deep parabolic reflector unit may be 
used to best advantage for general illumination in medium-sized or 
small sets, for which it is ideally suited because of its beam uniformity, 
wide angular spread, and high efficiency. The aperture spotlight 
is only rarely used, its field being limited, as has already been pointed 
out, to very special pattern lighting effects. 

Only the basic lighting units have been described, although many 
others are available which differ only slightly from those herein dis- 
cussed. It need not be said that they too will be found satisfactory, 
once their precise performance, modeling possibilities, and practical 
limitations have been investigated carefully. It is hoped that some 
of the principles here established will assist in the performance of 
this task. 





Summary. Color-temperatures of various carbon arcs have been calculated from 
spectral energy data. The dominant wavelength and per cent purity of each arc are 
given with reference to both "Average Daylight" and "Noon June Sunlight." 

It is pointed out that the color-temperatures of these carbon arc light-sources are of 
value in comparing them on a visual basis only. The effect of the radiant energy 
from the arcs upon any photosensitive medium other than the human eye, for example, 
photographic film, is very different from the visual impression. 

Spectral energy distribution curves of several carbon arc sources are published for 
the first time. 

In the last few months there have been a number of requests for 
the color-temperatures of various carbon arcs used in the motion 
picture industry. It is the purpose of this paper to provide such in- 
formation, but at the same time to call attention to the limited useful- 
ness of this method of expression as applied to the uses of the motion 
picture industry. 

In comparing light-sources on this basis, one must bear in mind 
that the effect of a light-source in terms of color- temperature is re- 
ferred to the sensitivity of the average human eye. From a visual 
standpoint, therefore, the color-temperatures of the carbon arcs may 
be very useful. 

It is very important to realize that color-temperatures will give very 
little and only approximate information as to the effect of light-sources 
upon photosensitive materials other than the human eye. The best 
criterion to use in evaluating the effect of a light-source upon photo- 
sensitive materials is the spectral distribution of the radiant energy 
of the source in question. It has been shown that one can calculate 
the effect of the radiant energy from any light-source upon any photo- 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received Sep- 
tember 17, 1937. 

** Research Laboratories, National Carbon Co., Inc., Cleveland, Ohio. 




7 V l\ \\ 



5000 j 6OOO 


FIG. 1. Comparative effect of light-sources upon photographic film and 
upon the eye: broken curves, sunshine; solid curves, 8-mm. motion picture 
studio arc, 37.5 volts, 40 amps. 

sensitive material if the spectral energy distribution of the light- 
source and the spectral sensitivity of the sensitive material are 
known. l We have shown in a previous paper the effect of the radiant 
energy from several carbon arcs upon a photographic film. 2 



FIG. 2. Spectral energy distribution of d-c. Suprex arc Positive crater 
radiation only (16-mm. Suprex positive, 5-mm. Suprex negative). One square 
represents 4 microvolts per sq. cm. at 10 feet. 

Fig. 1 gives the results of such calculations, showing the effect of 
sunlight and the light from 8-mm. National Motion Picture Studio 
carbons at 40 amperes and 37.5 volts upon photographic film and 
upon the average human eye. Comparison of the eye and film re- 







FIG. 3. Spectral energy distribution of d-c. Suprex arc Positive crater 
radiation only (7-mm. Suprex positive, 6-mm. Suprex negative). One square 
represents 4 microwatts per sq. cm. at 10 feet. 

sponse curves illustrates very well the very different results or effects 
of the radiant energy from a light-source upon two photosensitive 
media whose spectral sensitivities differ considerably from one an- 

FIG. 4. Spectral energy distribution of a-c. Suprex arc Positive crater 
radiation only (8-mm. Suprex positive, 7-mm. Suprex negative). One square 
represents 8 microwatts per sq. cm. at 10 feet. 

other. The futility of using color-temperature, which is a measure of 
the effect produced upon the eye, for the purpose of evaluating the 
effect of a light-source upon any other photosensitive material is thus 

April, 1938] 



The curves show also that the effects of the radiant energy from 
the sun and the 8-mm. National M. P. Studio carbons at 40 amperes 
and 37.5 volts are quite similar upon either the eye or supersensitive 
panchromatic film. 


FIG. 5. Spectral energy distribution of 9-mm. d-c. high-low reflecting 

projector arc Positive crater radiation only (9-mm. high-low positive, 

6 /ie-inch orotip c.c. negative). One square represents 20 microwatts per 
sq. cm. at 10 feet. 

The spectral energy distributions of 6-, 7-, and 8-mm. Suprex 
carbons, 9- and 11-mm. High-Intensity carbons, and 13.6-mm. 
Super High- Intensity carbons are given in Figs. 2 to 7, inclusive. 

The spectral energy distributions of low-intensity projector car- 


FIG. 6. Spectral energy distribution of d-c. high-intensity motion picture 
projector Positive crater radiation only (11-mm. h-i. projector positive, 
Vs-inch c.c. orotip negative). One square represents 20 microwatts per 
sq. cm. at 10 feet. 

bons, 3 National M. P. Studio carbons, 2i4>5)6 Rotary Spot carbons, 2 - 6 
16-mm. Sun Arc carbons, 2 ' 6 and 13.6-mm. High-Intensity carbons 8 
also used in the motion picture industry have been published pre- 
viously. Spectral energy distributions of sunlight and various other 
types of arcs will also be found in the literature. 7 * 12 



If one is interested only in the visual comparison of light-sources, 
color-temperatures are valuable and do serve to differentiate them 
from one another. The color-temperature of a source of light is 
defined as the absolute temperature at which a perfect black body 
must be operated in order to produce a color matching that of the 
source in question. Obviously this nomenclature can not be applied 
to any light-source whose color is widely different from that of a black 
body at any temperature. The various carbon arcs used in the 
motion picture industry do match black bodies closely enough so 
that color-temperatures give significant values. 

FIG. 7. Spectral energy distribution of d-c. high-intensity motion picture 
projector Positive crater radiation only (13.6-mm. super high-intensity 
positive, 7 /ie-inch extra-heavy copper-coated negative). One square repre- 
sents 20 microwatts per sq. cm. at 10 feet. 

In the determination of the color-temperatures given in this paper, 
the writers have followed the method outlined by D. B. Judd of the 
Bureau of Standards. 13 This method involves the calculation of the 
trichromatic coefficients of the color in question from its spectral 
energy distribution curve, according to standardized methods. 14 A 
comparison is then made between the location of the resulting point 
on a chromaticity diagram and the locus of black body colors. 

Through the use of a similar chromaticity diagram any light- 
source may be defined visually in terms of a fixed standard of com- 
parison, for example, average daylight, as standardized by the In- 
ternational Commission on Illumination in 1931 as Illuminant "C". 14 
The dominant wavelength and the per cent purity of the source in 

April, 1938] 




Color-Temperature of Carbon Arcs with Dominant Wavelength and Per Cent Purity 
Referred to Average Daylight 


Average Daylight 14 

11 -mm. H.I. Carbons 

8-mm. Suprex Carbons 

8-mm. Suprex Carbons 

16-mm. H.I. Carbons 

7-mm. Suprex Carbons 

l /j X 12 Rotary Spot Carbons 

6-mm. Suprex Carbons 

9-mm. H.I. Carbons 

7-mm. Suprex Carbons 

13.6-mm. Super H.I. Carbons 

13. 6-mm. H.I. Carbons 

6-mm. Suprex Carbons 

8-mm. Nat. M.P. Studio Carbons 

12-mm. Low-Intensity Carbons 

Current Voltage 

Tern- Dominant 
pera- Wavelength, Per 
ture Angstrom Cent 
K Units Purity 









































































Color-Temperature of Carbon Arcs with Dominant Wavelength and Per Cent Purity 
Referred to Noon June Sunlight at Springfield Lake, Ohio 


Noon June Sunlight 8 

8-mm. Nat. M.P. Studio Carbons 

12-mm. Low-Intensity Carbons 

6-mm. Suprex Carbons 

13.6-mm. Super H.I. Carbons 

13.6-mm. H.I. Carbons 

9-mm. H.I. Carbons 

7-mm. Suprex Carbons 

Ys X 12 Rotary Spot Carbons 

6-mm. Suprex Carbons 

16-mm. H.I. Carbons 

7-mm. Suprex Carbons 

8-mm. Suprex Carbons 

8-mm. Suprex Carbons 

11 -mm. H.I. Carbons 



Tern- Dominant 
pera- Wavelength, Per 
ture Angstrom Cent 
K Units Purity 








































































406 F. T. BOWDITCH AND A. C. DOWNES [J. S. M. P. E. 

question can then be determined with reference to this standard. 
The dominant wavelength is the wavelength of the spectral color 
that must be added to the standard to match the color of the source. 
The per cent purity of the light-source is its purity referred to this 
dominant wavelength. The higher the purity, the greater is the 
energy of the dominant wavelength that must be added to the stand- 
ard to match the color of the source in question. 

Table I gives the color-temperatures, dominant wavelengths, 
and per cent purities of various carbon arcs referred to average "day- 
light," and Table II gives the same values referred to noon June sun- 
light at Springfield Lake, Ohio. 8 

The dominant wavelengths and per cent purity values for any arc 
are not the same in the two tables because they are referred to differ- 
ent standards, "average daylight" in one case and a certain noon 
June sunlight in the other. If any other standard is chosen as a 
reference the dominant wavelengths and per cent purity values will 
be still different. 


1 JONES, L. A.: "Use of Artificial Illuminants in Motion Picture Studios," 
Trans. Soc. Mot. Pic. Eng. V (1921), No. 13, p. 74. 

2 BOWDITCH, F. T., AND DOWNES, A. C.: "Photographic Effectiveness of Car- 
bon Arc Studio Light-Sources," J. Soc. Mot. Pict. Eng., XXV (Nov., 1935), p. 

3 KALB, W. C.: "Present Trends in the Application of the Carbon Arc to the 
Motion Picture Industry," /. Soc. Mot. Pict. Eng., XXVII (Sept., 1936), p. 253. 

4 JOY, D. B., BOWDITCH, F. T., AND DOWNES, A. C.: "A New White-Flame 
Carbon for Photographic Light," /. Soc. Mot. Pict. Eng., XXII (Jan., 1934), p. 

* HANDLEY, C. W. : "Lighting for Technicolor Pictures," / Soc. Mot. Pict. 
Eng., XXV (Nov., 1935), p. 426. 

6 BOWDITCH, F. T., AND DOWNES, A. C. : "Radiant Energy Delivered on Mo- 
tion Picture Sets from Carbon Arc Studio Light Sources," /. Soc. Mot. Pict. Eng., 
XXV (Nov., 1935), p. 383. 

7 GREIDER, C. E., AND DOWNES, A. C.: "Sunlight Natural and Synthetic," 
Trans. III. Eng. Soc., XXV (1930), pp. 378-396. 

8 GREIDER, C. E., AND DOWNES, A. C.: "Physical Characteristics of Sunshine 
and Its Substitutes," Trans. III. Eng. Soc., XXVI (1931), pp. 561-571. 

9 GREIDER, C. E., AND DOWNES, A. C.: "The Carbon Arc as a Source of Arti- 
ficial Sunshine, Ultraviolet, and Other Radiation," Trans. III. Eng. Soc., XXVII 
(Sept., 1932), pp. 637-653. 

10 DORCAS, M. J.: "Ultraviolet Radiation in Industry," /. Ind. and Eng. 
Chem., XXII (Nov., 1930), pp. 1244-1246. 

11 GREIDER, C. E.: "Energy-Emission Data of Light-Sources for Photo- 


chemical Reactions," /. Ind. and Eng. Chem., XXIII (Ma}', 1931), pp. 508-511. 

11 KALB, W. C.: "Characteristics and Uses of the Carbon Arc," Elect. Eng., 53 
(Aug., 1934), p. 1173. 

18 JUDD, D. B.: "Estimation of Chromaticity Differences and Nearest Color- 
Temperature on the Standard 1931 I. C. I. Colorimetric Coordinate System," 
/. Opt. Soc. Amer., XXVI (Nov., 1936), pp. 421-426. 

14 "Handbook of Colorimetry," The Color Measurement Laboratory, Mass. 
Inst. of Technology, A. C. Hardy, Editor. 


MR. POPOVICI: How were the spectrograms made? 

MR. DOWNBS: The spectral energy distribution curves were determined with a 
quartz monochrometer by standardized methods originally outlined in the Bureau 
of Standards Scientific Paper No. 539, by Coblentz, Dorcas and Hughes. 

MR. RICHARDSON : In theaters we are interested most in the effect of the light 
upon the eye and the relation of the heat to the condition of the film. Have you 
made any such studies? All these arcs have light of various characteristics. 

MR. DOWNES: It is very doubtful whether anyone can visually distinguish dif- 
ferences in the lights from the various high-intensity arcs. 

MR. RICHARDSON: You think then that they would all be equally easy upon 
the eyes? 


MR. RICHARDSON : How about the heat developed at the aperture, and the ef- 
fect upon the film? 

MR. DOWNES: There is no light-source used for projection that does not de- 
velop a great deal of heat. The percentage of heat energy, if heat is defined as 
being the radiant energy of wavelengths longer than those visible to the human 
eye, is about 65 per cent of the total radiant energy. 

MR. SCHUMAKER: What is the maximum amperage in the super high-intensity 

MR. DOWNES: We have used as high as 195 amperes with 13.6-mm. carbons, 
and with the 16-mm. carbons 205 to 210 amperes. 

MR. RACKETT: I noted in two of the illustrations, differentiation between 
daylight and sunlight. What are the differences, and where and how are the 
measurements made? 

MR. DOWNES: Average daylight as used in this paper is the International Com- 
mission on Illumination's Illuminant C, which is a tungsten lamp operated at a 
certain absolute temperature with very carefully described chemical filters giving 
a color-temperature of 6500 K. That is a close approach to ordinary daylight, 
which is a mixture of sunlight and light from the sky. The sunshine value was 
from direct measurements of sunshine at Springfield Lake, Ohio, which is about 
fifteen miles south of Akron. 

MR. RACKETT: In the table, the column of deficiencies, in one case, revolved 
largely about the value of 5700 A. Was that the daylight comparison? 


MR. RACKBTT: The dominant wavelength is shown as 5700 A, and if my mem- 
ory is correct, 5500 A is about the center of the green band and about the peak 
sensitivity of the eye. 

408 F. T. BOWDITCH AND A. C. DOWNES [J. S. M. P. E. 

MR. JONES: 5550 A is the maximum. 

MR. RACKETT: That means, then, that the light is deficient in green? 

MR. DOWNES: The dominant wavelength is that of the radiant energy that 
must be added to the standard to produce the color match with the unknown light - 
source. It is not added to the unknown. 

MR. RACKETT: That is, the arcs are generally deficient in the green. In the 
table referring to sunlight the deficiency was lower in the spectrum, about 
4700 and 4800 A, from which we might draw the conclusion that June sunlight 
is richer in blue than the daylight that you used. 

MR. DOWNES: That is not correct; if it were, the color-temperatures of sun- 
light would be higher. The color-temperature of daylight is higher than that of 
sunlight, which means that there is more blue. For example, the color-tempera- 
ture of clear blue sky is estimated at 20,000 K; and as the temperature of any 
black-body source is increased, the dominant wavelength, or the wavelength of 
maximum energy, shifts toward the short end of the spectrum, so in skylight 
we have much more blue than in daylight, and in daylight there is more blue than 
in sunlight. 

MR. RACKETT: A point of interest in the curves is the sharp step in the blue 
region at about 3850 A. Am I correct in saying that that is largely due to the 
cyanogen band in the arc? 

MR. DOWNES: No cyanogen is liberated by the arc, although the so-called 
"cyanogen band" is present in the radiant energy from all carbon arcs. If cy- 
anogen is present in the arc stream, it is entirely burned and disappears. 

MR. KELLOGG: I believe I have read somewhere that for years astronomers 
wondered why there should be such strong cyanogen bands in sunlight, and 
found later that they could be accounted for on the assumption that some gas 
was doubly ionized. I thought that the idea that the bands were produced by 
cyanogen was abandoned. 

Is it not a little misleading, on the basis of color-temperature, to say that a 
light is deficient in the green? I thought that the color-temperature given meant 
that you began to lose in the green but lost still more in the blue. Is that right ? 

MR. DOWNES: It should be remembered that the dominant wavelength is the 
wavelength that must be added to the standard light-source to produce a match 
with the source under consideration. That does not mean that the divergence of 
the source in question from the standard is necessarily confined to the portion of 
the spectrum about the dominant wavelength. It is well known that any color 
can be matched with three primary colors, but sometimes in making a match on a 
colorimeter one of the three primaries must be added to the source being measured, 
which may be considered as a subtraction from the three primary colors in order 
to secure the match. 

MR. KELLOGG: Then the dominant wavelength may not bear much relation to 
color-temperature ? 

MR. DOWNES: No, it might not. 

MR. FRENCH: Can these carbons be burned under any conditions in the thea- 
ter at the prescribed voltage and amperage and maintain the same color-tem- 
perature, or are special burning conditions required? 

MR. DOWNES: The spectral energy distribution curves of the Suprex arc at, say, 
30 amperes and 40 amperes are not exactly alike, and the color-temperatures 


corresponding to those two amperages are somewhat different; but within the 
range of 30 to 40 amperes you would have a hard time visually to tell the differ- 
ence so far as the appearance of the light is concerned. 

MR. FRENCH: Does the color-temperature vary among carbons individually or 
in batches? 

MR. DOWNES: Our measurements on any given type of carbon have been re- 
markably uniform from one carbon to another as long as the arc voltage and cur- 
rent have been kept constant. 

MR. SHULTZ: In the motion picture theater, the eyes are generally dilated; 
they are in a darkened room looking at a screen having a brightness of 7 to 14 foot- 
lamberts. In sunlight the irises close, to compensate for the great quantity of 
light. Would not eye-strain become more apparent more rapidly in a theater 
wherein the eye must dilate more rapidly to compensate for changes? 

MR. WORSTELL: I believe the discomfort that the eye experiences in a theater 
is due to contrast rather than to the levels of illumination on the screen. The 
difference of brightness between the surroundings and the screen is in the dominant 
factor from the standpoint of visual comfort. Normal brightnesses outdoors, 
where the illumination may be as high as 8000 to 10,000 foot-candles, are not 
uncomfortable to the eyes because there is an absence of contrast. 

MR. RACKETT: The house illumination in many motion picture theaters is 
such that you see the surroundings of the theater with the low-intensity receptors, 
which I believe are the cones; whereas the screen illumination is considerably 
above the top levels of the cones, and you see the picture with the rods of the 
eye. The eye is therefore experiencing stimuli from sources having contrasting 
characteristics, not only in objects being seen, but in the receptive mechanism 
of the eye itself. The question is a little more complex than merely involving 
differences in color-temperature of the illuminants; probably the color- tempera- 
ture of the illuminant is a far less important factor in theaters than the conditions 
of the theaters themselves. 

MR. RICHARDSON: What effect does moisture have upon the characteristics 
of the arc? 

MR. DOWNES: After the first few seconds, none at all. If the carbons contain 
moisture it will cause "popping" at the arc until the moisture has been driven off, 
which may prolong slightly the time required for the arc to arrive at a steady con- 
dition. If care is taken to see that carbons are dry before trimming, no trouble 
of this kind will occur. 




H. A. SMITH** 

Summary. Within the past three years, two new types of stainless steel have been 
developed: (1) type 315 which contains approximately 18% chromium, 8% nickel, 
1.5% copper, and 1.5% molybdenum: and (2) a modification of type 316 (the usual 
18-8 S Mo) where the molybdenum content has been raised to from 3 to 4% molyb- 
denum. Considerable test data are now available for type 329, containing approxi- 
mately 27% chromium, 4.5% nickel, and 1.5% molybdenum. The latter steel shows 
promise in that pit-corrosion tendency is considerably reduced. Satisfactory welds 
may also be made with this type. From the corrosion-resisting standpoint, three other 
compositions are discussed: type 309, 24% chromium, 13% nickel; type 310, 25% 
chromium, 20% nickel; and type 446, 27% chromium. 

It is pointed out that a polished (No. 6) and a finely ground (No. 4) finish are 
more corrosion-resistant than a pickled finish, not only from the potential standpoint 
but due to the decreased possibility of their collecting foreign matter that will accelerate 
corrosive attack. 

In a previous paper 1 a short history was given of the development 
of stainless steels and of the application of certain types for structural 
equipment where corrosive conditions are mild. A more detailed 
resume was given of the application of the commoner types to proc- 
essing equipment. The present discussion will therefore be lim- 
ited to the presentation of experience obtained since 1934 with the 
commoner types of stainless steels and to experience obtained with 
special types. 


(a) Composition. Table I lists the alloy types and range of 
analysis obtained for each type. Types 304, 316, 329, and 446 were 
used before 1934; however, modifications have been made in 304, 
316, and 446 since 1934 to render them suitable for various applica- 
tions. Type 329 was known before this and was used principally in 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received October 
8, 1937. 

** Republic Steel Corp., Massillon, Ohio. 



o g 

3 r 1 -2 



Stainless Steel 
























2 1 





00 00 

T 1 












litable co 

I I 888888;8;88888 | 

t^ H^- O^ O5 O^ O^ CO ^O O^ O5 ^^ ^D 00 ^^ 


to totO^OtOOOOOOOOO j_ 



c3 ^^ ct5rtojrto3rtctJc3cjc3 ^ 

*"*2 ^5 * ^H ^H ^^ ^g ^^ ^q ^5 ^2 ^2 ^^ ,^2 W 

S ocq'oo 'ooodo'o ^^ 






<U <U r^ t-5 


pq "5 rb 


,^ 00 00 00 00 ^t CO 00 00 

w i j, ot i a o 2 ^ - j, < o 



[J. S. M. p. E. 

castings; but only recently has it been obtainable in strip, sheet, 
and plate form. Type 315 has been developed for improved service 
in special applications. Steels 329 and 316 with a high molybdenum 
content (high Mo) have been developed for use in corrosive media 
of a strongly acid nature. Type 304 has been modified and devel- 
oped for severe spinning operations. Type 446 is not suitable for 
welding but when modified is adaptable to arc, atomic hydrogen, and 
resistance welding. Type 347 is offered for welding because colum- 
bium does not burn out in the weld as readily as titanium, yet pre- 
vents precipitation of carbides and hence intergranular corrosion. 
Columbium has an additional advantage over titanium in that it 
does not lower the corrosion resistance of the sheet as much as ti- 
tanium. For the elimination of carbide precipitation in austenitic 
steels at the weld a low-carbon analysis (Type 304) is preferred; 
however, an 18-8 (Type 302) may be used where corrosive conditions 
are mild or where a proper anneal follows the welding operation. Of 
the steels listed only Types 430 and 303 may not be satisfactorily 

(b) Physical Properties. The range of physical properties of these 
stainless steels are given in Table II. 

Range of Physical Test Data for Various Types of Stainless Steel 




Elastic Limit, 
Lbs./Sq. In. 

Tensile Strength, Elongation, 
Lbs./Sq. In. (%) 2 Inches. 







18-8 B 



40,000/ 50,000 



18-8 FM 



35,000/ 45,000 

80.000/ 90,000 











40,000/ 60,000 






40,000/ 60,000 






45.000/ 60,000 



18-8 S Mo 









75,000/ 85,000 



18-8 S Cb 



35,000/ 45,000 

80,000/ 90,000 











45.000/ 60,000 

75.000/ 95,000 


Values shown in Table II compare the annealed as well as the 
extreme cold-rolled ranges of physical properties for most of the 
types listed. Types 329 and 446 (modified) may be given only 
shallow draws ; the austenitic type of chromium nickel steels may be 


deep-drawn without much difficulty. Type 430 takes only medium 
draws, unless annealed between drawing operations, but its drawing 
properties do not compare with those of the austenitic types. 

All these steels, in simple operations, are reasonably machinable 
when the proper technic of machining stainless is understood. Type 
303 is offered for intricate and delicate machining operations together 
with maximum corrosion resistance for a free-machining type. 

None of the alloys, with the possible exception of 329, is subject 
to hardening by heat-treatment. They may be hardened by cold 
working, i. e., by drawing or rolling. 

(c) Commercially Available Forms. Any of these analyses may 
be obtained in the form of hot-rolled, cold-drawn, turned, or center- 
less ground bar. They may be obtained in the form of cold-drawn 
or annealed wire and cold-heading rivets. Wood screws, machine 
screws, bolts and nuts, welded and seamless tubing, may be obtained 
in most of the analyses mentioned. 

Any of these types may be obtained in the form of hot and cold- 
rolled strip, sheets, and plates. The usual finishes obtainable on cold- 
rolled strip are : 

No. 1 (annealed and pickled) 

No. 2 (annealed, pickled, and rerolled) 

The usual finishes on sheets are : 

No. 1 (hot-rolled, annealed, and pickled) 

No. 2B (bright cold-rolled) 

No. 4 (commercially ground and polished) 

No. 6 (commercially ground, polished, and Tampico brushed) 

No. 7 (high-luster polish) 

No. 8 (mirror finish) 

The usual finishes on plates are : 

No. 1 (hot-rolled, annealed, and pickled) 

Annealed, pickled, and cold-rolled 
No. 4 (commercially ground and polished) 
No. 6 (commercially ground, polished, and Tampico brushed) 
No. 7 (high-luster polish) 
No. 8 (mirror finish) 


The corrosion resistance properties of stainless steels are due pri- 
marily to the relatively high chromium content of the alloys. The 
addition of nickel primarily influences the physical properties, al- 

414 H. A. SMITH [j. S. M. P. E. 

though, in general, it also favorably affects the corrosion resistance. 
The addition of molybdenum further influences favorably, in most 
instances, the corrosion resistance of such steels. Other additions 
to these basic analyses are made for obtaining variations in chemical 
and physical properties. 

It should continually be borne in mind that primarily the resis- 
tance of stainless steels to chemical attack is possible because of the 
passive film present on the surface of the alloys. Various additions 
to the steel and variations in surface treatment affect the stability of 
this film, and hence the corrosion resistance of the product. It has 
been fairly well established that this passive film is a very thin oxide - 
like film whose characteristics depend primarily upon the composition 
and physical structure of the underlying metal. 

Passivation treatment, i. e., chemical surface treatment, improves 
the corrosion resistance. The corrosion resistance is improved also 
by grinding and polishing, if properly done, i. e., by mechanical sur- 
face treatment. The higher the finish the more corrosion-resistant 
is the product. The various surface finishes listed in the order of de- 
creasing resistance to corrosive attack are : 

No. 8 (mirror finish) 

No. 6 (commercially ground, polished, and Tampico brushed) 

No. 4 (commercially ground and polished) 

No. 1 (hot-rolled, annealed, and pickled) 

This clearly shows that in finishes created by grinding and polishing, 
the smoother the finish the more highly corrosion-resistant is the 
steel. In the fabrication of equipment from steels with finishes 
above No. 1 it should be noted that if welding is employed the welds 
should be ground down and finished of! to the same finish as the sheet. 
Thus the entire surface presented to the action of the solution is as 
uniform in finish as possible. In making recommendations, the cor- 
rosion resistance, the alloy type, the finish, the relative costs of al- 
loy and finish, fabricating methods, and service conditions must all 
be considered. 

It may generally be stated, with perhaps minor exceptions, that 
corrosion resistance decreases with increase of temperature of the 
corrosive medium. It should be remembered, further, that while 
stainless steel is more resistant than most other metals or alloys to 
electrolytic attack resulting from concentration cells, it may be at- 
tacked where the stainless equipment is allowed to become exces- 
sively dirty in the presence of a corrosive medium. Processing equip- 


ment made of stainless steel should be cleaned periodically even 
though it is stainless, just as equipment made of other materials 
should also be cleaned. 


In general, the most severe corrosive conditions exist in various 
types of fixing baths. Here there is an acid condition with sulfites, 
thiosulfates, alums, and, occasionally, chlorides present as the active 
corrosive constituents in water solution. 

Developers may contain sulfites, sulfates, bromides, iodides, and 
borates as the more corrosive constituents. Such developers con- 
tain organic compounds which, together with basicity of the solution, 
render such solutions less corrosive than fixing baths. Acidic re- 
ducers and toning baths may contain chlorides, bromides, sulfides, 
nitrates, sulfates, dichromates, and other constituents. Such re- 
ducers and toning solutions are used for special work, and each bath 
of this type must be considered as a special problem. 

With the usual type of basic developer containing hydroquinone, 
elon, borax, and sodium sulfite, and with the ordinary fixing bath con- 
taining alum, acetic acid, sodium sulfite, and hypo, service experience 
and laboratory tests place the corrosion resistance of the various types 
of stainless in the order shown in Table III. In each listing in this 
table the type of steel at the top of the list is the most corrosion- 
resistant, and the one at the bottom is the least resistant to corrosive 


Corrosion Resistance of Various Types of Stainless Steel in Fixer and Developer 
(Listed in order of decreasing resistance to corrosive attack) 

Fixing Bath Developer 

316 (high Mo) 310 

309 329 

310 304 
316 302 
329 316 

446 316 (high Mo) 

304 446 

302 309 

Note: These listings are based upon results obtained on the steels when 
properly used, i. e., solutions were not allowed to dry or become concentrated in 
the equipment when not in use. 

416 H. A. SMITH [j. s. M. P. E. 

In the fixing bath column, all the steels, from the top of the list 
down through Type 329, have been found to give satisfactory service 
in fixing solutions. In the developer column, all the steels listed will 
give satisfactory service in the usual applications to the usual develop- 
ing solutions. It may thus be seen that the difference in perform- 
ance between the top and the bottom types listed under "devel- 
oper" is markedly less than the difference in performance between the 
top and the bottom types listed under "fixing bath." 

The superiority in fixing solutions of Type 316 (high Mo) is di- 
rectly reflected in the high molybdenum content as the passive film 
on such steels is exceptionally stable to acid solutions. The rela- 
tive superiority of the next two steels is reflected in their high chro- 
mium content. Type 316 has a lower chromium content, but this 
deficiency is made up by the addition of molybdenum. Type 329 
is of special interest, as it is less subject to the localized pitting type 
of corrosive attack than any of the other steels, except Types 316 and 
316 with high Mo content. When 329 is attacked, a general overall 
type of corrosion often occurs. 

In developer solutions, the 316 types of stainless are rated some- 
what lower than in fixing solutions, primarily because of the basicity 
of the former solutions. A passive film exceptionally stable in acid 
solutions is less so in definitely basic ones. 

Numerous tests have shown that fogging of film in the developer 
will not be encountered as the result of using stainless steel equip- 
ment for holding the developer. 


The importance of properly caring for stainless steel equip- 
ment when not in use should be stressed, as the majority of com- 
plaints received are from such causes. When equipment is to be 
idle for a period of time and it will be necessary to make up new solu- 
tions on processing again, all solutions should be drained and tanks 
and mechanical equipment flushed well with warm water before 
shutting down, to insure that solutions will not evaporate to dryness 
on the stainless steel. 


Type 316 has been used most extensively for tanks, trays, piping, 
and mechanical equipment exposed to the action of fixing solutions. 
As mentioned above, other types (329 and above) also give satis- 


factory service in such solutions. Rubber equipment has been re- 
placed by stainless steel in a number of instances where the acid con- 
ditions cause the rubber to exfoliate after some service. 

For developing solutions Types 302 and 304 have been most widely 
used and have proved to be quite serviceable. Any of the alloys in 
Table III may be applied, but Type 302 is most economical in first 

Individual storage of motion picture film presents a corrosion prob- 
lem satisfactorily met by the use of Types 430 and 302. In storage, 
film evolves small quantities of gases composed of a mixture of ni- 
trogen oxides, which attack other types of metal storage cases em- 
ployed. Such metal containers have been drawn satisfactorily from 
the 18-per cent chromium type of steel (Type 430) and 18-8 and thus 
far have resisted any corrosive attack. 

Ordinarily washing and drying equipment may safely be made of 
Type 302. 

Stainless steel has been used in the fabrication of photographic 
print dryers of which there are three main types : 

(A) Endless-belt type 

(B) Segmented-plate type 

(C) Rotating-drum type 

The essential requirement here is a high polish, and, in the plate 
and drum types, freedom from warping due to heat. In the first 
class of dryer, Type 304, cold-rolled and polished, is quite serviceable. 
In the plate and drum types, freedom from warping may be obtained 
by the use of Type 446 with a high polish. 

Wherever springs or clips are to be used, the application of Type 
316 (high Mo) is to be preferred. 

The use of stainless steel in the photographic industry has not been 
as general as it has been in other industries and as it should and un- 
doubtedly will become in photographic processing. This means that 
the experience of the steel producers in the application of their product 
is not as great as they should like. However, specific corrosion data 
are available for the various types of steel mentioned in this paper 
and will be furnished upon request. 

When possible, it is desirable that steel be subjected to tests under 
actual operating conditions, and to that end samples are furnished 
for testing, the results of which tests are of mutual benefit to the pro- 
ducer and consumer of stainless steels. 

418 H. A. SMITH [J. S. M. P. E. 

In recommending stainless steel for specific requirements it is es- 
pecially desirable that the steel user should cooperate as fully as pos- 
sible with the steel manufacturer in supplying fabricating and op- 
erating details so that the most suitable types of steel may be recom- 
mended. Only in this way can the greatest satisfaction from the use 
of stainless steel be expected. 


1 MITCHELL, W. M.: "Application of Stainless Steel in the Motion Picture 
Industry," /. Soc. Mot. Pict. Eng., XXIV (April, 1935), No. 4, p. 346. 


MR. CRABTREE: Under Type 316 there are two varieties with 2 and 4 per 
cent molybdenum. How long have you been supplying two types? 

MR. SMITH: The table designation does not mean 2 and 4; it means a range of 
molybdenum content that may be specified as 2 to 4 per cent. Up to about six 
months ago we had been supplying the steel with molybdenum between 2.25 and 
2.75 per cent. Our practice now is to supply a steel containing 2.75 to 3.25 per 
cent molybdenum. The steel with the modified molybdenum analysis contains 
molybdenum above 4 per cent and is not included in the regular 316 type of steel. 

MR. CRABTREE : Then we must specify the high content or the regular product? 

MR. SMITH: We prefer, depending upon the application, to vary the molyb- 
denum content. The reason for the range of analyses given for any of these 
steels is that by varying (within the range listed) the chromium or nickel or 
molybdenum, or any other specific addition, we may make a steel especially 
adaptable to the service conditions that must be met. 

MR. CRABTREE: Is the stainless steel business a custom business? In other 
words, if I ordered Type 316 2-per cent molybdenum with 7-A finish in, say, cold- 
rolled tubing do you carry it in stock, or do you have to make it to order? 

MR. SMITH: We keep in stock a large variety of analyses, both standard and 
special. We stock standard finishes on the more common analyses. Finishes No. 
1 and No. 2B are stocked for nearly all analyses. We can furnish, through our 
corporation, electrically resistance welded tubing of any of the analyses except 
the straight chrome types. 

MR CRABTREE: Getting back to the 316, you say that the range of content of 
molybdenum gives you a leeway. Can you control the content? 

MR. SMITH: The molybdenum content can be closely controlled. We have 
found from experience that, in certain applications where the details of the 
service are not known, Type 316 steel with 2.0-2.5 per cent of molybdenum has 
not been as satisfactory as it should be. For general applications we prefer this 
type of steel with molybdenum more in the center of the range, 3.0 per cent. 
When a customer can give us definite information with regard to the service con- 
ditions, we can apply the steel that will be most useful and most satisfactory for 
that application. 

MR. CRABTREE: In many cases we are looking for the most resistant metal we 


can find. In such cases what do we order? If we order 18-8 molybdenum, what 
do we get? 

MR. SMITH: Type 316 Enduro 18-8 S Mo containing 2.75-3.25 per cent of 

MR. CRABTREE: If we require the high molybdenum content, do you have to 
make it? 

MR. SMITH: We regularly stock 2.75 to 3.25 molybdenum steels. Values 
above that vary according to the application. On the modified Type 316 material 
there has been no Iron & Steel Institute standardization. We make it for special 
purposes up to 6 per cent molybdenum. 

MR. BRADLEY: The members of the Committee on Preservation of Film will 
remember the extensive discussion we had before we selected the containers for 
the storage of film in The National Archives, and finally selected the 18-8. We 
did not know at the time whether there was any particular difference in the finish. 
Samples were submitted to us in the pickle finish, satin, and the mirror finish. 
We got the pickle finish because it seemed a little cheaper. Now I learn that 
the pickle finish is not as resistant to oxides of nitrogen as the mirror finish. 
It is fortunate that I heard your paper this morning because I am just getting 
ready to order an additional quantity. You say the additional cost of the mirror 
finish would be justified in our problem? 

MR. SMITH: If you use the higher finish steels on welding the fabricated con- 
tainer, the weld must also be brought up to the same finish as the plate, otherwise 
the weld will be poor and the plate unsatisfactory. The cost is considerably 
higher, and I should think for your application that the pickle finish would be 
satisfactory. You have a condition where there are largely gases present, and 
not a water solution. 

MR. BRADLEY: I noticed in handling the cans that they are very easily soiled 
and finger-marked. What would you suggest as the best way of cleaning the 

MR. SMITH: If finger prints are fresh, they may be washed off with water, 
followed by benzene or some solvent for greases. Finger prints more difficult to 
remove may be cleaned off usually by treatment in 20 per cent nitric acid solu- 
tion at 130 F. If finger prints still persist, the container must be given a suitable 
light pickle. 

MR. CRABTREE: If a highly resistant sheet is not welded satisfactorily, or 
if the welding rod is not of the proper composition, corrosion may take place 
at the weld and ruin the whole apparatus? 

MR. SMITH: If the weld rod is not of the proper composition or if the welding 
is not properly done, corrosion may take place rapidly at the weld. In a highly 
resistant sheet that is satisfactorily welded with the proper rod there is little like- 
lihood of severe corrosion. 

In the case of individual film containers, which are merely compressed out of 
steel, you can use the higher finish quite satisfactorily, because the finish is not 
damaged if the drawing operations are carefully done. 

MR. BRADLEY: There is no welding at all; it is a matter of drawing the metal. 

MR. SMITH: Then I believe a higher finish might have an advantage. 

MR. BRADLEY: Our cabinets are made of the polished steel satin finish, I 
think. The containers are made of pickle finish. 

420 H. A. SMITH [J. S. M. P. E. 

MR. SMITH: The No. 8 finish comes at considerably higher cost than the 
No. 4 finish, which I believe should be quite satisfactory in case you want a higher 
finish than you are now using. 

MR. Cox: I should like to take slight exception to one statement of Mr. 
Smith's that nickel adds principally to the physical properties, and not chemical 
resistance. Most of us who are familiar with metals and alloys, particularly 
stainless steels, realize the importance of nickel from the standpoint of chemical 
resistance to corrosion. None of the materials we manufacture requires pas- 
sivating treatment, and I believe a little more information on the subject would 
be helpful. 

MR. CRABTREE: If you machine the metal or file it or weld it, or do something 
that will more or less break the skin, the resistivity of the metal is reduced, 
according to our experience. If you treat the metal with, say, hydrochloric acid, 
the passivation is removed. I would much prefer to work with a metal that does 
not depend upon passivation for its resistivity. We do not know when the metal 
is going to become non-passive and hence corrode. I have no objection to pas- 
sivation, but I certainly would not care to rely upon it. 

MR. SMITH: Mr. Crabtree is right. We should not depend upon a passivation 
treatment for creating a surface on a steel so that it may be applied in a corrosive 
agent that would otherwise call for a higher grade of steel. That is one reason 
why we apply the molybdenum type of stainless steel in so many instances; 
because we find it has a more stable passive film and when damaged will reform 
itself more readily than on the other types of steel. Passivation should be done 
in 20-per cent nitric acid for 10-15 minutes at 150F. Such treatment is prin- 
cipally useful in cleaning up a stainless surface that may have picked up mild 
steel contamination which is likely to cause pitting and rusting. 

MR. Cox: Were the tests to which you referred made with passivated stain- 
less or with materials, say, taken right from stock? 

MR. SMITH: They were made with materials taken from stock and freshly 
pickled, in the standard pickle of nitric hydrofluoric acid. They were not pas- 

In answer to the comment about nickel, my statement was that nickel added 
largely to the physical properties and that it also helped the chemical properties. 
However, we find in gases containing sulfur, and in some solutions containing 
sulfites, that nickel sometimes is not as advantageous as a steel containing higher 
chromium content. That is, an 18-8 would be less preferable than a steel con- 
taining 25 per cent chrome without the addition of the nickel. 

MR. CRABTREE: With regard to acid fixing solutions containing silver, of the 
commercially available metals or alloys the 18-8 molybdenum is the most resis- 
tive material we have been able to find to date. That is why I am interested 
in knowing something more definite about the molybdenum content. I am 
anxious to get a steel with more molybdenum for more resistivity. That is why 
I wish Mr. Smith would specify the content more definitely rather than merely 
as "high" and "low." 

MR. SMITH: You have been using Type 316 steels with molybdenum contents 
of about 2.5 and 2.80 per cent. The table in the paper is the kind specified by 
the Iron & Steel Institute and shows a range of composition. We preferred to show 
this sort of table rather than specific compositions because from our experience 


we are able to apply variations with the range of the analysis for special uses. 

MR. HUBBARD: In the majority of cases where stainless steel is used for these 
purposes, we can not order in quantities big enough to have the content made 
up to suit our particular needs. What is the best thing to order that you have 
in stock? 

MR. SMITH: The Enduro 18-8 S Mo, Type 316, 2.75-3.25 per cent molybdenum. 
There is, of course, some slight variation in analysis from heat to heat, but from 
the corrosion -resistance standpoint such small variations are not significant. 
When you want the most resistant type of steel, we ordinarily furnish this steel 
unless there is some specific requirement as to resistance, in which case we may 
apply the special molybdenum steel, which has not yet received an Iron & Steel 
Institute number. 


Summary. A system of air-conditioning is described that employs lithium 
chloride for independently controlling both the relative humidity and the dry-bulb 
temperature of air. It is used both for comfort air-conditioning and for treating air 
for industrial processing work. 

Lithium chloride is one of the most hygroscopic of inorganic compounds, and the 
aqueous solution has the property of absorbing moisture from, or adding moisture to, 
the air, depending upon the vapor pressure difference between the air and the solution. 
From this it is seen that, by properly controlling the concentration and temperature, 
the lithium chloride solution is capable of either dehumidifying or humidifying the 
air, depending upon the requirements. The air is cooled or warmed when passed 
over an aqueous solution of lithium chloride, depending upon whether the solution 
is cooler or warmer than the air. Further cooling or warming of the air when desired 
is attained by using an after-cooling or after-heating coil. 

The cycle of air-conditioning is explained and illustrations of an air-conditioning 
unit are shown. The application of the system to a typical problem of interest to 
motion picture engineers is discussed and illustrated by means of a schematic flow 
diagram. Operating data for full-load and for less than full-load conditions show 
low cost of operation and efficiencies equally as high when operating either at maxi- 
mum load or at less than maximum load. Washing, deodorizing, and neutralizing 
bacteria from the air by contact with lithium chloride are important factors where 
pure clean air is desired. 

Lithium chloride is used in air-conditioning for controlling both the 
relative humidity and the dry -bulb temperature of air independently. 
It is used for comfort air-conditioning and also for treating air for 
industrial processing work. 

The importance of an air-conditioning system that provides in- 
dependent control of relative humidity and dry -bulb temperature is 
shown in Fig. 1, which shows the variation for August, 1936, in Day- 
ton, Ohio, in tons of refrigeration required for moisture removal and 
tons of refrigeration required for sensible cooling per 1000 cu. ft. per 
minute of outside air to comfort conditions. These curves show that 
during a considerable portion of the time very little temperature re- 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received October 
5, 1937. 

** Surface Combustion Corporation, Toledo, Ohio. 




duction is required at the same time that a large amount of moisture 
should be removed. With this system, the moisture removal and the 
temperature reduction are two entirely independent operations, and 
the user can control either at will. 


29 31 

FIG. 1. Refrigeration required for moisture removal and 
sensible cooling, per 1000 cu. ft. per min. of outside air, to 
comfort conditions (Aug., 1936). 

Lithium chloride is one of the most hygroscopic of inorganic com- 
pounds, and the aqueous solution has the property of absorbing mois- 
ture from, or adding moisture to, the air, depending upon the vapor- 
pressure difference between the air and the solution. The equilib- 
rium vapor-pressure of the solution is dependent upon the tempera- 
ture and the concentration of the solution, so tnat by controlling these 

424 G. A. KELLEY [J. S. M. P. E. 

items properly the solution is capable of either dehumidifying or 
humidifying the air, depending upon the requirements. By cooling 
the solution below the temperature of the air for the dehumidifying 
operation, the air can be cooled at the same time the moisture is re- 

The viscosity of lithium chloride solution is higher than that of 
water, but low enough to permit the use of ordinary pipe sizes for 
the circulation of the solution. 

The cycle for the lithium chloride system of air-conditioning is 
quite simple. For summer operation the air to be conditioned is 
brought into contact with lithium chloride solution at a vapor-pres- 
sure less than that of the air, so that moisture is absorbed from the 
air by the solution due to the vapor-pressure difference. In absorbing 
moisture from the air a certain amount of heat is generated, which is 
the heat of condensation of the water-vapor in the lithium chloride 
solution. The temperature of the air in passing over the solution 
tends to rise, due to the loss of water- vapor; but the temperature 
rise of the air can be kept down by precooling the solution to a tem- 
perature below that of the air. In most comfort applications, the 
solution is precooled by cooling water in a solution-cooler, so that 
the air in passing through the air conditioner is cooled as well as 

The solution passing to the air conditioner is continuously cooled 
in a solution-cooler with city, well, refrigerated, or cooling-tower 
water. The quantity of water required is a function of the water 
temperature and the total heat, either sensible or latent or both, 
removed from the air by the solution. Where water is scarce or ex- 
pensive, an evaporative cooler can be used to cool the solution directly, 
without the use of a cooling- tower. In this case, the water require- 
ment is only that required for make-up purposes, and amounts to a 
fraction of a gallon per minute per ton of equivalent refrigeration. 

The lithium chloride solution is weakened by the absorption of 
water, and a small percentage of the solution is passed to the solution 
conditioner where it is concentrated by driving off the water absorbed 
by the main body of solution. The solution conditioner consists of a 
contact surface in which the weak solution is heated with ten-pound 
steam to a temperature at which the vapor pressure of the solution 
is considerably higher than that of outside air and contacted with out- 
side air for the purpose of removing water from the solution. 
Three-pound steam call be used with slight increase in steam con- 



sumption. The steam can be produced with a small low-pressure, 
gas-fired boiler, automatically controlled for summer and winter 
operation, which requires no attention by the user. The moist air that 
is discharged to the outdoors also has a cooling effect upon the hot 
solution as it passes through the concentrator. The warm concen- 
trated solution returns to the sump and mixes with the main body of 
the solution to maintain the solution at the desired concentration. 
Considerable operating economies are effected in this system because 

2800 CFM ea'DB 70 WB 55% 

FIG. 2. Flow diagram showing application of lithium chloride air-condition- 
ing system to a typical industrial drying problem. 

the solution conditioner is under automatic control and operates only 
as required to remove the water absorbed from the air, with the re- 
sult that the fuel consumption is in direct proportion to the water 
removed. With the equipment operating at 50 per cent of maximum 
moisture-removal load, the fuel consumption is approximately 50 per 
cent of that required for maximum load. 

Lithium chloride solution absorbs, to a very large degree, most 
of the odorous materials in the air. The odors absorbed from the 
air by the solution are driven off in the concentrator and discharged 
to the outdoors so that the solution, after concentration, is odorless. 



[J. S. M. P. E. 

In this way there is a continuous cycle in which odors are absorbed 
from the air in the air conditioner and then removed from the solution 
in the solution conditioner. The greater portion of the dust and dirt 
in the air is removed and bacteria neutralized when the air is washed 
with lithium chloride. 

For winter operation, the air to be conditioned is preheated and 
brought into contact with lithium chloride solution at a vapor-pres- 
sure higher than that of the air, so that moisture is added to the air by 

FIG. 3. Standard self-contained lithium chloride air-conditioning 


the solution due to the vapor-pressure difference. There is prac- 
tically no change in the dry-bulb temperature of the air in passing 
through the air conditioner under winter operation. Heating or 
cooling coils can be added to the air conditioner when necessary to 
effect any heating or cooling of the air beyond that effected by passing 
the air over the solution. 

This cycle permits independent removal of moisture from the air, 
since the moisture content of the air leaving the air conditioner is 
dependent upon the temperature and concentration of the solution, 
which is under automatic control, so that the concentrator operates 
as required to maintain the proper strength of solution. 



Fig. 2 is a schematic flow diagram showing the application of the 
lithium chloride air-conditioning system to a typical industrial drying 
problem in which the drying takes place along the adiabatic line ex- 
cept for extraneous losses and gains. In this particular problem 250 
cu. ft. per min. of fresh air was admitted to the air conditioner in 
order to keep the room under pressure, and the drying process was 
such that 2750 cu. ft. per min. of air was introduced from the air 
conditioner to the drying room. The operating data given in Fig. 2 

FIG. 4. Application of lithium chloride system of air-conditioning to a typical 


indicate that the drying cost of this system is extremely small com- 
pared with that of the more usual drying methods. However, in 
many cases, the savings in operating cost are not the major considera- 
tion governing the choice of a drying system. Accurate control of 
dry -bulb temperature and relative humidity of air is often of great 
importance in maintaining uniform quality and uniform drying rate 
of product. This is attained with the lithium chloride system because 
it is possible to achieve independently modulated control of dry-bulb 
temperature and relative humidity of the air within very close limits. 
The system should be of considerable interest to engineers because of 

428 G. A. KELLEY [J. S. M.'P. E. 

the wide range of drying conditions possible at low operating costs. 
Equipment for maintaining process rooms at drying conditions such 
as 73 dry-bulb and 18 per cent relative humidity, and 100 dry- 
bulb and 10 per cent relative humidity is now in operation. Fig. 3 
is a standard self-contained lithium chloride air-conditioning unit 
that will treat up to 2750 cu. ft. per min. of air. The circulating 
fan and all equipment are within the cabinet. 

Fig. 4 shows the application of the lithium chloride system of air- 
conditioning to a typical theater. Design conditions for this theater 
are 750 occupants, with an internal sensible heat gain of 188,000 
Btu per hour, an internal latent heat gain of 132,000 Btu per hour. 
Forty-five hundred cu. ft. per min. of fresh air and 6500 cu. ft. per 
min. of recirculated air pass through the lithium chloride air con- 
ditioner, where it is dehumidified, deodorized, the bacteria neutralized, 
and the air passed to an after-cooling coil where sensible heat is re- 
moved before discharging the conditioned air to the theater. The 
total computed sensible heat load is 19.8 tons and the total latent 
heat load 19.1 tons, which is a total load of 38.9 tons. Inside design 
conditions are 80 dry-bulb, 50 per cent relative humidity, with out- 
side conditions of 92 dry-bulb, 104 grains of moisture per pound of 
dry air. In this application, sensible cooling of the air is obtained 
with 55 water. Mechanical refrigeration or re-evaporation is used 
for sensible cooling in cases where the water temperature is too high. 

Summarizing the operating data shown in Fig. 4 for maximum 
summer design operation, it is seen that the operating cost is very 
small : 

60 gal. per min. of 55 water at 12 per 1000 gal. = $0.432/hr. 

410 pounds per hour of 10-pound steam at 50$ per 1000 pounds = 0.205/hr. 
2.2 B.H.P. for solution pump at 2 l /rf per kw. = 0.055/hr. 


or a total of 39.8 tons of refrigerating effect at $0.0175 per ton- 

Assuming the average load to be 70 per cent of full load, the operating data 
would be: 

30 gal. per min. of 55 water at 12 per 1000 gal. = $0.216/hr. 

280 pounds per hour of 10-pound steam at 50 per 1000 pounds = 0.140/hr. 
2.2 B.H.P. for solution pump at 2 l /rf per kw. = 0.055/hr. 

$0.41 1/hr. 

or a total of 28 tons of refrigerating effect at $0.0147 per ton- 


The better-than-proportional decrease in operating cost while 
operating at part load with this system is obtained because energy 
is used for direct and independent removal of the latent load and the 
sensible cooling load. 

In many public places, such as theaters, the purifying and deodor- 
izing, and neutralizing bacteria are as important as the savings in 
operating cost. Lithium chloride, because of its antiseptic and 
purifying characteristics, offers a means of purifying the air as well 
as of reducing the dry-bulb temperature and moisture content. 

For winter operation, the total computed heat loss for heating and 
humidifying is 385,000 Btu per hour, so that the same boiler is used 
for winter operation as for summer operation of the air-conditioning 
equipment. The same equipment is used for winter operation as for 
summer operation with the exception that heating coils are used in- 
stead of cooling coils. 

To summarize, the lithium chloride system of air-conditioning is 
designed for year-round operation to perform the following functions : 

(7) Dehumidifying in the summer-time. 

(2) Cooling in the summer-time. 

(5) Heating in the winter-time. 

(4) Humidifying in the winter-time. 

(5) Circulating. 
() Cleaning. 

(7) Washing. 

(8) Deodorizing. 

(9) Neutralizing bacteria in the air. 


MR. CRABTREE: It always seemed to me to be a wasteful procedure to cool 
air and then to warm it again in order to dehumidify it. Your system would ap- 
pear to be the ideal one for theaters that are likely to overdemonstrate to the pub- 
lic that they have cooling systems, and produce temperature differences between 
the inside and the outside air of 30 degrees. 

I noticed that considerable mixing of air and solutions was indicated on the 
chart. Is the procedure fixed, or does the janitor have to juggle the controls? 

MR. KELLEY: The mixing of the solution at the start is practically balanced. 
We have a solution-cooler on one side and a solution-heater on the other side, 
and the pressure drop through the two is set so that the mixing is done auto- 
matically. An orifice or valve is set at the factory and the rate is never changed. 

The air is mixed once. The hand damper is set when the equipment is put into 
operation and is not changed unless the operator wishes to vary the air flow. 

MR. CRABTREE : What are the engineering advantages of lithium chloride as 
compared with those of other desiccating agents such as calcium chloride and 
potassium carbonate solutions? 

430 G. A. KELLBY [j. s. M. p. E. 

MR. KELLEY: I am not familiar with the potassium carbonate, but as for com- 
paring it with calcium chloride, the vapor-pressure of lithium chloride is con- 
siderably lower, so it is possible to get a much lower dew-point of air leaving the 
lithium chloride solution. 

Another advantage of lithium chloride over calcium chloride is the freezing 
point. At the concentrations we use, lithium chloride freezes at about 45, 
which of course is lower than the temperatures at which we operate. The freezing 
point of calcium chloride is considerably higher. 

MR. CRABTREE : It is interesting that the lithium chloride solution is so effective 
in absorbing the gases and the odors from the air. Is it more effective than, say, 
water? We all know that activated charcoal, and substances with very large sur- 
faces, are very active in removing or adsorbing gases and odors. Why is the 
lithium chloride particularly effective? 

MR. KELLEY: In most of our air washing we recirculate the water. With 
lithium chloride we have a continuous cycle. We absorb the moisture and the 
odors in the air conditioner, and then at the same time on the other cycle we have 
the solution conditioner where the moisture and odors are driven off the solution. 
We have some evidence that lithium chloride has a greater absorbing capacity for 
odors than has water. 

The continuous cycle, such as we use, is probably very important in removing 
odors from the air. One demonstration of odor removal in one of our installations 
this summer was very startling. Some oily rags were set afire accidentally in the 
fresh-air duct, and the oily smoke passed through the lithium chloride air-condi- 
tioning equipment and then was discharged into the various conditioned rooms. 
Of course, everybody became alarmed immediately and ran into the rooms. 
There was no trace of odor at all in the air-conditioned rooms. 

MR. ROBERTS: How effective or applicable is this system for film drying? 
Would it be applicable in a fairly close system where we do not have to add much 
outside air, where we clean the air once and keep it fairly clean? 

MR. KELLEY: It is very applicable to such a drying problem. The system is 
very applicable where very little fresh air is required, and has a very low operating 
cost. The operating conditions are easily within limits of the equipment. 

MR. CARVER: Is this the only company that makes this system? 

MR. KELLEY: Yes. There is one other company that made the equipment 
for shoe drying conveyors, as part of the conveyor equipment. 

MR. CRABTREE: I presume the cost of the lithium chloride is insignificant 
compared with that of the mechanical equipment. I have to admire the man 
who had the courage to proceed with a commercial process such as this, using 
what chemists have always regarded as a rather rare chemical. 

MR. KELLEY: Commercial lithium chloride, of course, is not as expensive as 
the chemically pure. However, its cost is somewhat of an item, although I 
believe that the reason is, as you say, that up to the present it has been quite a 
rare chemical and there has not been a great use for it. The price seems to be 
dropping now with increased use, though we wish it did not cost quite as much as 
it does. 

MR. MILLNER: Is there any danger of corrosion of the ducts or the apparatus 
by the solution? 

MR. KELLEY: Of course, no solution gets in the ducts, and as long as we select 


the materials properly in the various parts of the equipment we are able to avoid 
corrosion. If we were to pick the metals without carefully choosing them, we 
should have trouble due to corrosion. 

MR. PATTERSON: Do I understand correctly that the efficiency of this method 
is a function of vapor-pressure difference between the air treated and the lithium 


MR. PATTERSON: It occurs to me then, particularly for motion picture theater 
conditioning and where it would seem desirable to introduce as much outside air 
as possible, that it would be economical to treat a comparatively larger quantity of 
outside air with lithium chloride solution. In that case you would be treating 
a high moisture content air with a low vapor-pressure solution and the wide vapor- 
pressure difference will give a high efficiency of moisture removal. You would be 
operating the machine at greater efficiency by treating all outside air rather than 
by using a mixture of outside air and recirculated air, and at the same time pro- 
vide more fresh air per person. 

MR. KELLEY: That is quite correct. It would be that much easier to remove 
the moisture and the system would be operating at a little higher efficiency. 
However, the lithium chloride, which is very good at deodorizing, also has an 
antiseptic value, and we feel that we do not need all the fresh air that would be 
needed otherwise. By recirculating the air, purifying it, and neutralizing the 
bacteria, so much fresh air is not required. 


J. C. FOX** 

Summary. Die-castings are defined as castings made by forcing molten metal 
into a metallic mold or die. The alloy most generally used is of the zinc base type, 
having a tensile strength of approximately 40,000 Ibs. per sq. inch. For photographic 
appliances, the alloys of lower specific gravity are more desirable. Aluminum base 
alloys are used more extensively in photographic appliances for that reason. Physical 
properties of various aluminum die-casting alloys are given. 

Since low specific gravity is of prime importance in castings used for photographic 
appliances, the development of the process of die-casting the lightest of all commercial 
metals, magnesium, is of particular interest to motion picture engineers. Mag- 
nesium is one-third as heavy as aluminum, and magnesium die-castings are now 
being used wherever light weight is important. Physical properties of magnesium 
die-castings are given. Reference is also made to die-casting brass and German 
silver, a recent development. 

It is doubtful whether there is a metal fabricating process that has 
contributed so greatly to the economic quantity production of metal 
parts as has the die-casting process, defined as the production of metal 
castings by forcing fluid metal into a metallic mold or die under ex- 
ternal pressure. 

Die-castings have assumed great importance in many branches of 
industry and have been instrumental in the development of world- 
wide markets for many domestic appliances. Some of the reasons for 
the extensive use of this method of -production are : 

CO High-speed production of strong, accurately finished castings at relatively 

low cost. 

(2) Production of castings of close dimensional limits. 
(5) Production of sound castings with surfaces so smooth as to require very 

little preparation for plating, painting, and other finishes. 

(4) Production of rigid castings of very thin wall section. 

(5) Production of castings of a wide variety of shape and sizes, from a very 
small part weighed in grams to a large automotive grille casting weighing 
about 27 Ibs. 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received October 
12, 1937. 

** Doehler Die Casting Co., Toledo, Ohio. 


(6) Ease of machinability. 

(7) Good corrosion resistance. 

() A wide selection of alloys of varying physical and chemical properties. 


Die-castings have played a prominent role in the automotive, ma- 
chinery, household utility, electrical, radio, hardware, business ma- 
chine, and other industries. Without the economy and speed of 
production of die-casting, it is safe to state that many industries 


1. One-piece zinc die-casting of motion picture 

would not be able to hold the wide markets created. What is true 
of developments in these industries is true of the photographic ap- 
pliance industry. Zinc base, aluminum base, and brass die-castings 
are being extensively employed as parts for still and motion picture 
cameras and projectors. One sound projector, recently developed 
contains a total of thiry-five die-cast parts, of which 17 parts are of 
aluminum base alloys, 13 parts of zinc base alloy, and 5 of brass alloy 

Zinc Alloys. Motion picture manufacturers have been quick to 
recognize and take advantage of the many economies inherent in 


J. C. Fox 

[J. S. M. P. E. 

zinc alloy die-castings. The intricate designs of most motion picture 
parts present obvious applications for this modern metal and process. 
The projector housing illustrated in Fig. 1 is a one-piece die-casting 
instead of an assembly of many parts, without the many machining 
and assembly operations that other methods of manufacture would 
entail. Several popular low-priced cameras and projectors have 
been constructed of zinc alloy die-castings. In no other metal and by 
no other process could these cases have been produced at comparative 
low cost and durability without sacrificing appearance. Other die- 
castings in several of these units are the shutter mechanism frame 

within the camera, and the 
shutter wheel housing main 
frame and film feeder housing 
on the projector. 

The die-cast film reel for a 16- 
mm. commercial camera shown 
in Fig. 2 presents an interesting 
study in complexity. There is 
no other way that this part could 
be produced with the same 
qualities and economies of a 

Aluminum Alloys. Because of 
their characteristic properties, es- 
pecially that of low specific gravity, aluminum die-castings are ex- 
tensively employed in photographic appliances, as in other portable 
equipment where weight is an important factor. Figs. 3 and 4 illus- 
trate various assemblies and parts made of aluminum die-castings. 
Such castings have been used for camera cases for both still and mo- 
tion picture cameras; frames for projectors and mechanisms; hous- 
ings for shutter, clamp, fan, lamp, and lens mechanisms; brackets 
for take-up and rewind mechanisms; lens mounts; range finders; 
tripod tables and parts; shutter parts; and slide blocks and various 
other mechanical parts. 

Fig. 3 illustrates an assembly of six aluminum die-castings for a 
projector. The use of light aluminum alloys with extremely thin- 
walled sections makes an ideal combination for small light-weight 
pocket cameras. Fig. 4 illustrates a large collection of miscellaneous 
aluminum die-castings used in photographic equipment. 

Brass Die-Cast Parts. Brass die-cast parts in photographic equip- 

FIG. 2. Die-casting of motion pic- 
ture film core. 

April, 1938] 




ment have been used especially where their high strength 
resistance to wearing have been required. 

Magnesium Base Alloys. Of great interest to the photographic 
appliance manufacturer is the development during the past few 
years of die-casting magnesium base alloys. Although magnesium 
alloy castings are not especially new, the die of magnesium is com- 
paratively new. The development of a suitable casting machine and 
process for casting this type of metal and alloy has greatly expanded 
its use commerciallv. 

FIG. 3. 

Assembly of aluminum die-castings in 

As designers and engineers become more familiar with the proper- 
ties of magnesium and its alloys, numerous uses will be found for 
them. The greatest use of these alloys will perhaps be in the ex- 
tension of applications, because of its ultra-lightness. A study of the 
mass-strength ratios of magnesium to other metals will clearly indi- 
cate that this material can be employed advantageously for many 
structural parts. Magnesium has not yet received the attention it 
deserves, as there are undoubtedly many uses to which strong, light 
alloys can be put, resulting in higher working speeds and lower run- 
ning costs. 


J. C. Fox 

[J. S. M. P. E. 

The most outstanding characteristic of magnesium is its extreme 
lightness. The average specific gravity of magnesium alloys is 1.80 
or a weight per cu. in. of 0.065 Ib. A saving of one-third of the weight 
when magnesium replaces aluminum, and three-fourths of the weight 
where steel is replaced, can be affected by using magnesium alloy die- 

Another important quality of magnesium and its alloys is in ma- 
chinability. Magnesium is machined faster and better than any 

FIG . 4 . Miscellaneous aluminum die-castings used in photographic appliances . 

other metal. Magnesium possesses good physical properties and has 
demonstrated its stability and resistance to atmospheric exposure in a 
large number of applications extending over a number of years. 


Zinc die-castings are finished in a variety of ways, both for decora- 
tion and protection against tarnishing. There are, however, many 
applications in which zinc die-castings are used in the natural or "as 
cast" state without surface treatment. Such parts may be used for 
long periods of time without any effect upon their utility. Surface 
oxidation of zinc castings may be greatly inhibited by the application 


of a short dip in certain solutions of alkali metal chromates or di- 

The majority of zinc die-castings used in photographic equipment 
are enameled, japanned, lacquered, or finished in a combinaton of 
electroplate and enamel. They may be enameled and high-lighted 
sections may be simply polished and the exposed zinc protected 
against tarnishing by the application of a coat of clear varnish. 

Realistic wood-grain designs may be readily applied to die-castings, 
as well as other novel finishes of a great variety of color, design, and 
texture. Crackle, crystal, opalescent, wrinkle, cobweb, suede, metal 
luster, and other effects are possible. Many novel and attractive 
patterns may be obtained by combining some of these finishes. For 
instance, a combination of wrinkle or crackle finish topped with a 
flat top-coat of solid color gives a leather-like appearance. Alu- 
minum, gold, or bronze enamels can be used in connection with a 
wrinkle finish to produce an attractive effect. A number of parts 
are finished in bright lustrous chromium for decorative trim. Dull 
or "Buter" chromium may also be used. 

Aluminum. 'Aluminum die-castings are most generally finished by 
polishing, enameling, or anodically oxidizing and coloring. Polished 
aluminum die-castings retain their luster for long periods of time under 
ordinary atmospheric conditions. It is because of this property that 
most aluminum applications are finished simply by polishing and 
buffing to a high luster. Aluminum can be finished with the same 
organic and novelty finishes described under zinc. 

Magnesium. Magnesium alloy die-casting surfaces are protected 
against tarnishing by a suitable organic coating. Protection and 
decoration can be obtained together by selecting the proper paint 
materials and paint schedules. A definite procedure has been es- 
tablished which, if followed, will secure lasting protection to the mag- 
nesium surfaces. Once the surface has been cleaned properly, chemi- 
cally treated, and a good baked primary coat applied, any of the 
regular finishing enamels and novelty effects can be used to produce 
final finishes having both sales appeal and utility. 


The three major requirements of the die-casting process are : 

(1) The use of a casting machine to hold and operate a die. 

(2) A properly designed and constructed die. 
(5) A suitable alloy. 


J. C. Fox 

[J. S. M. P. E. 

Successful die-casting depends upon the careful coordination of all 
the details of each of these factors, and requires an organization 
trained in all branches of engineering and metallurgy to affect this 

To make die-castings render their full value, close cooperation be- 
tween the designer and the die-casting engineer is essential. Early 




FIG. 5. Principle of zinc-base alloy die -casting machine. 

contact between these persons, especially during the formative stage 
of the design of a part, will result in applying die-castings to the best 
advantage. The die-casting engineer can advise how to design a 
part that will minimize die costs and production difficulties; how 
to save weight where weight is not necessary; what alloy will best 
meet the functional requirements of the part ; how to minimize plat- 
ing and finishing costs; and how to design to reduce assembly costs. 
These factors must be considered to derive the best results for both 
producer and consumer. 

April, 1938] 



The buyer must be assured that his source of supply of die-castings 
is thoroughly dependable. The alloys used must be subjected to 
chemical and metallurgical control in the preparation and in casting 
operations, and kept within the limits set by standard specifications. 
The control of accuracy and uniformity of castings during the casting 
operation and the elimination of defects by rigid inspection methods 
are, of course, also necessary. 

Strength, accuracy, and uniformity are fundamental factors of 
quality of die-castings. By proper design, careful selection, and con- 

FIG. 6. Principle of aluminum die-casting machine. 

trol of the alloy, strength is maintained. Accuracy is obtained by 
careful calculation of shrinkage and by accurate die construction. 
Uniformity depends upon complete coordination of all factors and 
upon the elimination of variables. 


Die-casting machines can be divided conveniently into three major 
divisions: (1) plunger type machines; (2) air type machines; (3) 
High-pressure hydraulic ram types. 

Die-Casting Methods. The lower melting point metals, tin, lead, 
and zinc base alloys, are usually cast in what is termed a plunger type 

440 J. C. Fox [J. S. M. P. E. 

machine. In principle, this machine consists of a melting pot in 
which is submerged a plunger pump (Fig. 5). The inlet orifice to the 
pump is beneath the level of the metal, and the outlet orifice is con- 
nected to the sprue or gate of the die. When the plunger is raised, 
molten metal fills the pump cylinder, and when the plunger is de- 
pressed, the molten metal is forced into the die cavity. 

A piston operated by compressed air is usually employed to actuate 
the metal pump plunger. A system of toggle linkages connected 
between the air cylinder and the metal plunger is generally employed 
to increase the pressure at the end of the stroke. 

Aluminum Die-Casting. Aluminum die-castings can not be made 
commercially on the plunger type machine just described for the 
following reasons : 

Aluminum has a tendency to dissolve or wash away the ferrous 
metal of which the metal pump is constructed, thus quickly wearing 
away the piston fit and resulting in a loss of pressure; also, oxides of 
aluminum are formed, which tend to impart to the piston a binding 
action and quickly make it inoperative. 

The method employed in casting aluminum alloys is as follows; 
Submerged in the molten aluminum alloy is a specially shaped cast- 
iron container known to the industry as a "gooseneck" (Fig. 6). This 
gooseneck might be compared to a tea-kettle suspended by the handle. 
A pipe from a compressed-air supply is connected to the lid opening. 
The gooseneck is mechanically suspended on a system of links so that 
it can be submerged sufficiently below the surface of the melting pot 
to bring the orifice of the spout below the surface of the molten metal 
in the pot. 

The linkage is so arranged that the gooseneck can then be raised so 
that the spout can be brought out of the molten metal, and moved 
into contact with the sprue or gate of the die, at which point it is me- 
chanically locked in place. 

The operations are as follows: The gooseneck is lowered into the 
molten alloy. The compressed-air connection to the lid is opened to 
the atmosphere by means of a valve, and the air-pressure is shut off 
by means of another valve. This allows the molten metal to flow 
into the gooseneck through the spout. The gooseneck is then me- 
chanically lifted from the melting pot, allowing the surplus metal to 
drain from the spout. It is then tilted backward slightly to prevent 
dripping and the continuation of the stroke brings the spout into 
contact with the die orifice where it is securely locked. 


The valve leading to the atmosphere is then closed and the valve 
from the compressed-air supply is opened. Approximately 500 Ibs. 
per square-inch of air-pressure is applied to force the molten metal 
into the die impression. The compressed-air supply is then shut-off 
and the valve to the atmosphere is opened, after which the gooseneck 
is allowed again to drop back into the melt to receive a fresh charge. 

All the operations are generally performed automatically in syn- 
cronism with the rest of the die-casting machine. The other prin- 
cipal function of a die-casting machine consists in opening and closing 
the dies, pulling the cores, and ejecting the finished castings. 

Brass Die-Casting. Neither the plunger nor the gooseneck types 
of machine can be used for die-casting brass alloys. Brass when die- 
cast is handled in small charges, and is forced into the die at unusual 
speed and pressure (Fig. 7) ; whereas zinc is cast at pressures up to 
1800 Ibs. per square-inch, and aluminum 500 Ibs. per square-inch. 
Brass is cast at pressures up to 10,000 Ibs. per square-inch. 

The Design of Dies. The design of the dies is the first and most 
important step in the manufacture of die-castings. A die that is im- 
properly designed can not produce satisfactory castings, and fre- 
quently a die so constructed can not be corrected. There are many 
vital factors to be considered by the die designer before deciding upon 
the construction of any die, some of which are as follows : 

The location of the gate or metal inlet is of vital importance to the 
user of die-castings. The gate location is decided upon not only 
from the standpoint of casting production, but also from the user's 
viewpoint of the finished casting. The designer of a die is much 
better able to locate the gate properly if he is thoroughly familiar 
with the service application of the die-cast part. 

Since all die-casting dies are made of steel it is evident that methods 
must be provided for the exit of air in the die cavity when it is dis- 
placed by the metal entering into the die. Failure to provide proper 
vents in the die will naturally result in unusual porous castings. 
Many methods of venting are employed. The size, contour, and 
section of the casting determine the method of venting that is most 

Proper means must be provided in every die for rapidly ejecting 
the casting from the die after the casting operation. This is usually 
accomplished by means of ejector pins. In the process of ejecting 
the casting, the ejector pins make a slight impression in the casting. 
These ejector pin marks are frequently quite visible. It is the task 


J. C. Fox 

[J. S. M. P. E. 

of the die designer to place these ejector pins in such position on the 
casting that they eject the casting rapidly without distortion, and, at 
the same time, the ejector pin marks must not be objectionable from 
the user's viewpoint. 



FIG. 7. Principle of brass die-casting machine. 

Die Steels. The difficulties met in die-casting practice increase and 
multiply with advancing melting and casting temperatures of the 
alloys used. The life of die-casting dies varies with increase of melt- 
ing and casting temperatures. For instance, dies for lead and zinc 
base alloys with casting temperatures not exceeding 800F may be 


made of ordinary machine steel (S. A. E. 1040) and without heat treat- 
ment are capable of long use before becoming unserviceable. 

For aluminum alloys with casting temperatures around 1300F, the 
greater heat absorption makes it necessary to use a high-grade alloy 
steel composition for the dies. Furthermore, proper heat treatment of 
the alloy steel die is essential in order to obtain a life of 100,000 casting 
cycles before replacement is necessary. 

Brasses and bronzes having still higher casting temperatures and 
greater heat absorption values naturally present more difficulties 
than aluminum. 


The important qualities of a suitable die-casting alloy may be 
listed as follows : 

(1) Physical properties: strength, ductility, hardness, etc. 

(2) Permanence of properties and dimensions in service. 
(5) Fluidity. 

(4) Strength and ductility at elevated temperatures of solidification. 

(5) Solidification range. 

(6) Low liquid and solid shrinkage. 

(7) Machinability. 

(<?) Polishing, plating, and finishing properties. 
(9) Corrosion resistance. 
(10) Weight and cost. 

The alloys used in modern die-casting practice may be classified 
into six main groups, namely : 

CO Tin base alloys: tin alloyed with antimony, copper, and lead. 

(2) Lead base alloys: lead alloyed with tin and antimony. 

(5) Zinc base alloys: zinc alloyed with aluminum, copper, and magnesium. 

(4) Aluminum base alloys: aluminum alloyed with copper, silicon, nickel, and 

(5) Copper base alloys: copper alloyed with zinc, tin, silicon, aluminum, nickel, 

(6) Magnesium base alloys: Magnesium alloyed with aluminum, manganese, 
and silicon. 

A description of the general properties and applications of typical 
alloys of each group follows : 

Tin and Lead Base Alloys. Tin and lead base alloys represent only 
a small part of the total die-casting production, although for some 
uses these alloys are indispensable. Because of their corrosion re- 
sistance, tin base alloys find considerable application in parts for 

444 J. C. Fox [J. S. M. P. E. 

soda fountains, milking machines, syrup pumps, dental appliances, 
and surgical instruments. 

Lead base alloys are usually employed where a cheap non-corrosive 
metal is required, and where strength and hardness and other me- 
chanical properties are unimportant. Parts that must withstand the 


Chemical Composition of Zinc Base Alloys 

A. vS. T. M. Alloy XXI A. S. T. M. Alloy XXIII 

S. A. E. 921 S. A. E. 903 

(Per Cent) (Per Cent) 

Copper 2.50-3.00 Max. 0.10 

Aluminum 3 . 50-4 .30 3 . 50-4 . 30 

Magnesium 0.02-0.05 0.03-0.05 

Iron Max.-0.10 Max. -0.10 

Lead Max.-0.007 Max. -0.007 

Cadmium Max.-0 . 005 Max.-0 . 005 

Tin Max.-0.001 Max.-0.001 

Zinc (99.99 + % purity) Remainder Remainder 

action of strong mineral acids, as in fire extinguishers or other chemi- 
cal apparatus, are produced in lead base alloys. X-ray equipment 
employs lead die-castings, because of the resistance lead offers to the 
passage of x-rays. 


Alloy XXI Alloy XXIII 

Tensile strength (Ibs./sq. in.) 48,000 40,000 

Charpy impact strength (ft. Ibs.) 18 20 

Elongation (% in 2") 5 5 

Brinell hardness 83 74 

Compressive strength (Ibs./sq. in.) 93,000 60,000 

Shearing strength (Ibs./sq. in.) 45,800 31,000 

Melting point 715F 718F 

Specific gravity 6.7 6.6 

Thermal conductivity (cal./sec./cm. 3 /C) 0.25 0.27 

Thermal expansion (per C) 27.7 X 10 - 155,000 

Zinc Base Alloys. Of all the types of alloys used in die-casting 
the zinc base alloys are used in the majority of cases, chiefly because 
of their low cost, their ease of casting, their physical properties, the 
stability of these properties, and their excellent finishing qualities. 

In Table I are listed the zinc:base alloys in most common use today. 
These conform to specifications issued by the American Society for 
Testing Materials (B-86-33-T), and the Society of Automotive 
Engineers (Nos. 921 and 903). 

April, 1938] 



<5 CT-H 


o o 


x rt 

D ^D 


O O 


CO <M 


aT o o t-n 

<M 1-1 


J?^ SE Jvj 5 


T oo co ^ 
S ^ ^ 


g>>00000000 ^coff^fN 

..2iOiOOiOT-icOtO(M g o'OOO 

i^co'op'oq'dddddo 1^88^ 

^ w o o x x x x x x* rt 25 ^ ^ 

Tn o ^O ^o rt rt ctf c3 rt c3 *"^ 5J ^^ T ~^ 

8 I 


Ig^ggsg^s I 

!3 CC1 HI Co 



i "o" rt' S rt* X ' ^' X ' ^ ^ -^ 

| '8 

O "ti iO 

o co 

CO rH 

. O 00 


^ ^1 <N T-H T^ IO b- 

* d O O Q O "? 

x o x x x x x x 



O G 

446 J. C. Fox [J. S. M. P. E. 

The American Society for Testing Materials and the Society of 
Automotive Engineers exact the following minima of physical prop- 
erties from these alloys : 

A. S. T. M. A. S. T. M. 

Alloy XXI Alloy XXIII 

S. A. E. 921 S. A. E. 903 

Minimum tensile strength (Ibs./sq. in.) 44,000 35,000 

Minimum charpy impact strength (ft. Ibs.) 6 12 

Minimum elongation 2 . 3 

In Table II physical properties and constants capable of these two 
alloys as cast from die-cast test-bars made in accordance with 
A. S. T. M. specifications are given. 


Nominal Composition of Standard Copper Base Die-Casting Alloys 
I II in 

(Doler Brass) (Doler Brastil) (Doler Nickel Brass) 

Copper 65 81.0 44 

Zinc 34 14.75 40 

Silicon 1 4.25 
Nickel 14 
Manganese 2 

Tensile strength 70,000 85,000 70,000 

Elongation (% in 2") 25 10 0.10 

Yield point 40,000 60,000 55,000 

Brinell hardness 120 170 140 

Specific gravity 8 . 50 8 . 25 8 . 55 

Aluminum Base Alloys. Aluminum base alloys are used chiefly 
for their characteristic properties, which may be summed up as 
follows : 

(7) Low specific gravity. 

(2) High thermal and electrical conductivity. 

(5) Corrosion resistance. 

(4) Freedom from any dimensional changes. 

(5) Stability of properties at subnormal temperatures. 

(6) Ability to hold a polished luster for long periods of time. 

It is because of these properties, especially lightness and corrosion 
resistance, that aluminum die-castings are used extensively for port- 
able equipment such as vacuum sweepers, typewriters, photographic 
equipment, and many others. 

The standard aluminum base alloys used for die-castings are es- 
sentially those covered by A. S. T. M. specifications B-85-33-T. 


These alloys, which are listed in Table III with their physical 
properties, may be classified into three main groups : 

(1) Copper- aluminum alloys. 

(2) Copper-silicon-aluminum alloys. 

(3) Silicon-aluminum alloys. 

Copper Base Alloys. Pressure die-casting copper base alloys (brass) 
is relatively new, but progress is proceeding steadily under a definite 
program of research and development (see Table IV). 

The salient qualities of brass die-castings, great strength and cor- 
rosion resistance, demand attention from design engineers. The die- 
casting of brass alloys opens to industry a source of engineering parts 
of exceptionally great strength and accuracy, intricacy, and stability 
not readily obtained by other means of fabrication. It is difficult to 
limit the potential market of materials having such a combination of 
properties. Brass die-castings have already found use in a number of 
motion picture camera and projector assemblies. 


Magnesium Base Alloy 

Doler Mag. No. 2 Doler Mag. No. 6 Doler Mag. No. 10 

Aluminum 2.00 6.00 10.00 

Manganese 0.10 0.10 0.10 

Tensile strength (Ibs./sq. in.) 23,500 24,000 28,000 

Yield point (Ibs./sq. in.) 15,500 16,000 21,000 

Impact strength (ft. Ibs.) 6.0 3.0 1.0 

Elongation % in 2" 

8.0 3.0 1.0 

Brinell hardness 42 52 65 

Weight (per cu. in.) 0.064 0.065 0.066 

Melting point (F) 1175 1155 1100 

Magnesium Base Alloy. The standard magnesium base alloy 
used for die-casting has the chemical composition and physical prop- 
erties shown in Table V. 


MR. CRABTREE : Can you die-cast the stainless steels? 

MR. Fox: Not as yet. The development of the die-casting art has proceeded 
from the casting of alloys of low melting points to those of higher ones. At first 
tin and lead base alloys, with melting points up to 700 F were cast. Then 
followed in order zinc base alloys with melting points up to 800 F, aluminum 
base alloys with melting points up to 1200F, and copper base alloys with melting 
points up to 1600 F. The highest melting point materials, such as the bronzes, 

448 J. C. Fox 

cast iron, and steel will be the next tried, but shall await development of better 
die materials. 

MR. CRABTREE: When casting magnesium how do you prevent oxidation? 

MR. Fox : We use an inert gas atmosphere over the molten magnesium . Sulfur 
dioxide gas is most generally used for this purpose. 

MR. PALMER: What are the relative costs of magnesium and aluminum? 

MR. Fox: The specific gravity of magnesium is about 1.8 as against 2.8 for 
the lightest aluminum alloy. Although the cost per pound of magnesium is higher 
than that of aluminum, the difference in the densities makes the cost per volume 
of magnesium less than that of aluminum. 

MR. PALMER: Are there any practical difficulties in using magnesium for die- 
casting which would make it unsatisfactory? 

MR. Fox: Aside from the care necessary to prevent oxidation of the melt, 
there are no more difficulties in die-casting magnesium than in die-casting alumi- 
num alloys. There are certain precautions that the supplier or consumer may take 
in machining magnesium which concern the inflammability of finely divided 
magnesium. Magnesium does not take fire until the melting point is reached. 
The fine chips take fire readily because they can not conduct the heat away fast 
enough. It is only when large accumulations of chips are carelessly left in the 
open that any fire hazard is found. 

MR. CRABTREE: But the metal oxidizes very readily in a moist atmosphere, 
does it not? 

MR. Fox: That is right. 

MR. CRABTREE: And you have to protect it with a suitable lacquer. 

MR. Fox: Yes. It is important that proper procedure and proper materials 
be used in finishing magnesium castings. A properly cleaned casting, and a 
suitable baked primer coating followed by the color coatings is the usual procedure. 
One concern makes a complete vacuum sweeper that is so finished and has proved 
its practicability in every respect. 

MR. SWARTZ: Is the magnesium more brittle than the aluminum? 

MR. Fox: Magnesium is about the same as aluminum in that respect. Of 
course, the compositions of both aluminum and magnesium alloys control this 
property. A 10 per cent aluminum -magnesium is comparable with the stand- 
ard aluminum casting alloys, such as the commercial No. 12 alloy. With lower 
aluminum content, in magnesium alloys, the ductility is increased. 



Summary. The phenomenon of adsorption is discussed. Some of the properties 
of the solid adsorbent, activated alumina, are given, including a dynamic characteristic 
curve. This dynamic characteristic is utilized industrially to dry air and gases to 
dewpoints as low as -76C (0.0004 grain per cu. ft.; 0.0009 milligram per liter). 
Apparatus utilizing activated alumina in this way is described and pictured. 

Uses of the system include drying controlled atmospheres and bottled gases, and in 
chemical processing when water-vapor would promote corrosion or adversely affect 
the processes. Compressed-air lines are kept free of water to prevent freeze-up in 
winter or spoilage of work. 

In industrial and comfort air-conditioning, comparatively large quantities of 
partially dried air are required. A continuous dehumidifier to meet these require- 
ments is described and illustrated. 

It is pointed out that performance is a function of machine design as well as of the 
fundamental characteristics of the solid adsorbent used. Depending upon the 
factors of first cost and economy of operation, a wide range of performance may be 
obtained. A curve shows the present-day characteristic of a line of machines that is 
commercially available. 

The phenomenon of adsorption is not thoroughly understood. It 
can, however, be described in sufficiently accurate terms to form a 
useful mental image of it and its practical application. 

Attention will be confined to the adsorption of gases by solids, 
since water-vapor is essentially a gas and its removal from mixtures 
of other gases is becoming an increasingly important process. 

It has been found that an affinity exists between certain surfaces 
and the molecules of certain gases that causes gas to cling to the sur- 
face. We may say in effect that the gas is then condensed to a liquid, 
and a heat equivalent to the latent heat is given off. Furthermore, 
the application of heat releases this bond, and frees the adsorbent of 
its adsorbed gas. 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received Septem- 
ber 9, 1937. 

** Vice-President and General Manager, Pittsburgh Lectrodryer Corp., Pitts- 
burgh, Pa. 




[J. S. M. P. E. 

Since the adsorbability of different gases by any given adsorbent is 
roughly proportional to the boiling point of the gases, this process of 
adsorption affords a means of separating certain gases from others in 
a mixture. Adsorption is probably a universal phenomenon, but 
certain substances can be so prepared as to show adsorption to a 
marked degree. 

To be of commercial value such an adsorbent must have a high 
removal efficiency, a high adsorptive capacity, and freedom from 










The Variation of vapor pressure 
with percentage adsorption from air at 
30C and 90 per cent relative humidity 
for 8-14 mesh activated alumina by the 
dynamic method. 




10 15 


FIG. 1. 

Per Cent Water Adsorbed on Basis of Dry Weight 
of Acrivated Alumina 

Characteristic curve showing adsorption efficiency of activated 

chemical reactions with the gases to be treated. It must also be 
available in large quantities at a price that can be warranted by the 
results attained through its use. Such a solid adsorbent is activated 

Chemically this material is largely aluminum oxide. Physically 
its structure is such that a large adsorptive surface is available per 
pound of material. Since this material exhibits marked capacity to 
adsorb water- vapor from air and gases, it has wide commercial ap- 

Gases passed through beds of this material can be made extremely 
dry. A dry ness equivalent to dewpoints below 76 C has been ob- 

April, 1938] 



tained. Fig. 1 shows the characteristic curve of the material under 
the conditions stated on the chart. 

Taking advantage of this characteristic of activated alumina, air 
for the manufacture of liquid air is dried to extreme dryness in equip- 
ment similar to that shown in Fig. 2. Apparatus of the same type is 
used for drying carbon dioxide, ethylene, carbon monoxide, hydro- 
gen, oxygen, and, in fact, practically all the commercially known gases. 

More and more chemical and 
metallurgical processes are being 
conducted in controlled atmos- 
pheres. Sheet steel and strip in 
the past always had to be pickled 
after annealing to remove the 
scale formed through oxidation 
of the metal surface. This prac- 
tice is now being discontinued. 
Great tonnages of sheet steel and 
strip are now coming out of the 
annealing furnaces with the same 
bright finish that they had when 
they entered. This is accom- 
plished by the control of the 
chemical nature of the gases 
with which the heated steel 
comes in contact. Often this 
control requires reduction of the 
water-vapor content. Apparatus 
shown in Fig. 3 is utilized for 
drying the controlled atmos- 
pheres for annealing metals, to 
prevent the decarburization of high-carbon steel in heat treatment; 
and to prevent discoloration in the drawing furnaces. 

Protective atmospheres are used also in the manufacture of paints 
and varnishes and other chemical products. This control again often 
involves the reduction of the water-vapor content. 

Compressed air is freed from water-vapor in apparatus of this type 
so that it will not precipitate or freeze pipe lines or blow condensed 
water onto the work in spray booths or when testing or merely blow- 
ing the dust off parts in the process of manufacture. 

We believe, however, that the motion picture industry will be 

FIG. 2. Apparatus for drying air 
and other gases at high pressure. 



[J. S. M. P. E. 

more interested in applications involving comparatively large quan- 
tities of air partially dehumidified for the control of humidity in air- 
conditioning and in drying operations. In this instance factors of 
economy dictate a different method of attack in the utilization of a 
solid adsorbent. 

Where continuous drying of the air and gas is desired, two or 
more adsorbers are used so that one mav be reactivated before the 

FIG. 3. Apparatus for drying controlled annealing atmospheres. 

other has played out. The air-conditioning equipment we shall now 
discuss has eight adsorbers. Four of these adsorbers are on duty 
removing moisture from the air-stream; three are on reactivation; 
and a fourth is on the cooling or purge stage just prior to putting it 
back into service. As each adsorber plays out, it is put on reactiva- 
tion and a fresh adsorber cut into the circuit. This is done by means 
of a rotating central distributor, the only moving part in the appara- 
tus except the motor and blowers. Reactivation of the spent ad- 
sorbers is conducted continuously by means of a stream of heated 

April, 1938] 



air, and the purge serves to cool the heated adsorbers prior to putting 
them back into adsorption. 

The performance of such apparatus is a function of machine de- 
sign based upon the properties of activated alumina. A characteris- 
tic curve of these machines as now built is shown in Fig. 4. 

This "Lectrodryer," as it is known in the trade, is essentially an 
apparatus for the reduction of the water- vapor content of the air. 
Air passed through this machine by its self-contained blower comes 
out dehumidified in accordance with the characteristic curve shown 

-42 3 

3 $ 

I 1 

la la 

2345 67 

Grains per Cu. Ft. Input Air 

31 42 49 56 61 66 

Dewpolnt of Input Air (F) 

FIG. 4. Characteristic curve showing dryer performance. 

above. In its passage through the machine this air is raised to a 
temperature of about 150F. This is caused in part by the heat of 
adsorption but includes also some of the heat stored in the adsorbers 
during the reactivation stage. 

In some applications this heated air is of use; for instance, where 
warm, dry air is needed for a drying operation. In other cases, as in 
air-conditioning, the heat must be removed before the dry air is used 
for dehumidification. This is then accomplished by after-coolers 
in the outlet stream, where either well water or city water or auxiliary 
refrigeration may be used. 

In such a machine the heat energy required for reactivation will 
amount from 2 l /z to 4 times the latent heat of the water adsorbed, 



[J. S. M. P. E. 

depending upon the size of the machine and the moisture content of 
the input air. 

Removal of water-vapor alone is often the most important single 
problem in industrial air-conditioning. A dehumidifier in the room 
to dry the air by recirculating it through the machine or an arrange- 
ment to feed partially recirculated and some fresh air through the 
dehumidifier, is utilized. 

Fig. 5 shows a very simple case involving reduction of the water- 
vapor content in a storage room. No attempt is made to control the 

FIG. 5. Dryer for lowering water-vapor content of storage 


temperature. Materials stored here are highly hygroscopic, and 
large quantities were ruined by the weather annually until this ap- 
paratus was installed. Note the simplicity. The machine is placed 
right in the same room or in an adjoining room if more convenient. 
A single duct leads the dry air into the room and another duct re- 
turns it to the machine for redrying. The humidistat in the room 
turns on the machine automatically when the moisture content of the 
room rises above a predetermined figure, and shuts it off when the 
room is dry enough. In summer air-conditioning applications this 
dehumidifying unit may be combined with circulation, filtering, and 
sensible heat removal. 

April, 1938] 



Drying of most materials can be accomplished by simply using 
heated air. Some products, however, are affected by temperature 
so that it is not desirable to heat the material above a certain range. 
In these instances the humid summer air contains so much moisture 
that sufficiently low relative humidity for properly drying these 
products can not be obtained. This is either because the process 
is unduly prolonged or sufficient dryness can not be obtained during 

FIG. 6. Large apparatus for prevention of condensation of 
moisture due to evaporative cooling effect of solvents. 

periods of high atmospheric humidity. In such instances the dehumi- 
dification of atmospheric air fed to drying cabinets, tray dryers, and 
drying lofts forms a very interesting solution to this problem. It is 
particularly important in such cases to remember that the normal 
output of an adsorption dehumidifier is warm, dry air. In consider- 
ing the overall cost of dehumidified air due credit should be given to 
the heat energy available in the air dried this way. 

In the manufacture of film base, water-vapor may be deposited in 
the material by cooling due to solvent evaporation. This may be 

456 G. L. SIMPSON [J. s. M. P. E. 

avoided by using dehumidified air in the atmosphere surrounding the 
"wheels." Fig. 6 shows a machine used to condition over 15,000 
feet of air per minute for just such an application. 

In other applications the evaporation of solvents in the manu- 
facture of mica insulation caused local cooling which precipitated water 
from the atmosphere ; this was corrected by the installation of equip- 
ment for controlling the dewpoint of the air in the room in which the 
material is manufactured. 

Hundreds of activated alumina dehumidifiers have been installed. 
They range in operating pressures up to as high as 3500 pounds per 
square-inch and in capacities in excess of 15,000 feet per minute. 
Frankly, we do not know for what purposes some of these machines 
are used. Engineers are investigating more and more the effects 
that water-vapor may be having on their processes or on their product. 


MR. CARVER: In the diagram showing the rotary air switch you indicated 
that the air was passed through one of the driers in the opposite direction to get 
it ready. How is that done? 

MR. SIMPSON: We did not draw the details in the lower part of the drawing. 
The distributor up to the mid-point is composed of three sections: the main 
stream, the reactivating stream, and a little stream that feeds air into the ad- 
sorber getting ready to go back into service. The latter purges the beds of the 
products of combustion and gas-burning equipment, and serves to take some of 
the moisture out of the bed. The stream of heated air is then returned to the 
reactivating stream. 

MR. BRADLEY: Is there any self-contained unit suitable and practicable for 
studios and rooms having no outside outlet? I am thinking of human beings 
rather than of goods. 

MR. SIMPSON: The machines made by my company are not yet developed for 
household or office installations. 

MR. BRADLEY: Are such machines on the market? 

MR. SIMPSON: I believe there are. 

MR . CRABTREE : How frequently must the alumina be renewed ? Is it poisoned 
in any way by use? 

MJR. SIMPSON: Anything that will paste up the structure or shut it off will 
poison it sugar dusts, etc. With the inlets properly filtered, these machines have 
run for four and one-half years on the same charges. It is simply a question of 
keeping the dust and dirt out of the machine. The usual chemicals that will 
not poison men will not poison this material. Nitrous oxide and some of the 
other oxides of nitrogen, for instance, have some effect upon the material, but the 
quantities available in the normal processes are not enough to have any effect 
commercially upon the application of the machine. As a matter of fact, we are 
using just that property to remove nitrous and other oxides of nitrogen from cer- 


tain controlled atmospheres where otherwise we would get very distinct corrosion 
of pipe lines due to the moisture content. 

MR. CARVER: Must you use gas for reactivation, or can you use steam? 
Is it cheaper to use gas? 

MR. SIMPSON: We want a stream of heated air. If it is cheaper to use gas, 
that equipment is available. If it is cheaper to use steam, that equipment is also 
available. In some cases we use electricity, where the fire hazard and the lack 
of steam make electric operation desirable. Many of these machines are running 
on gas, which is a particularly effective fuel and in many cases provides a dis- 
tinctly inexpensive way to operate the equipment. 

MR. BRADLEY: Is it possible to remove oxides of nitrogen in any great quan- 
tity? I am thinking about film fires, where the fumes and smoke represent a 
definite hazard to human health and life. Would it be possible to direct these 
fumes through such a machine and neutralize their toxic effects in large quantities? 

MR. SIMPSON: That is a thought I certainly shall follow up. The concentra- 
tions with which we have dealt have been a matter of 1 and 2 per cent, cumula- 
tive, however, in pipe lines, and very corrosive. We shall certainly look into the 
possibility of removing much larger concentrations from air streams. 


During the Conventions of the Society, symposiums on new motion picture appara- 
tus and materials are held in which various manufacturers of equipment describe and 
demonstrate their new products and developments. Some of this equipment is de- 
scribed in the following pages; the remainder will be published in subsequent issues 
of the Journal. 


Although sound-level meters have been commercially available for some time, 
it is only within the past year that they have attained their present high degree of 
popularity. One of the main reasons for this sudden acceptance by industry is, 
doubtless, the availability of new and improved models combining convenience of 
operation, low weight, and, in some cases, low price. 

Probably few industries have as many important uses for a sound-level meter 
as the motion picture industry. Noise meters, as they were formerly called, have 
long been used for measuring the noise-levels in studios and theaters, the sounds 
made by various mechanical and electrical devices such as ventilators, cameras, 
projectors, arc lights, etc., and for checking the volume of reproduction and the 
background noise-level from reproducing systems. Recent improvements in 
microphones, however, have made possible sound-level meters having reasonably 
smooth frequency characteristics, so that such instruments, unlike the earlier 
noise meters, are suitable not only for measuring complex noises, but are also 
quite satisfactory for many kinds of single-frequency measurements. Naturally, 
this has expanded the usefulness of the sound-level meter to include measure- 
ments of the overall frequency response of reproducing systems and variations in 
frequency response throughout a theater or auditorium. 1 

One of the newest sound-level meters is the General Radio Type 759-^4, which 
incorporates many features hitherto unavailable in even the most expensive instru- 
ments (Fig. 1). Aside from meeting the tentative specifications of the American 
Standards Association, the design of this new instrument stresses portability and 
convenience of operation, which characteristics are of utmost importance to the 
user. Among its many features are a non-directional sound-cell microphone, a 
high-gain stabilized amplifier, the absence of all battery adjustments, a practically 
linear decibel meter, and a simple system for resetting the calibration. 

To mention the features in more detail, the sound-ceil microphone provides a 
rugged and sensitive sound pick-up device with a smooth, nearly flat frequency 

* Presented at the Fall, 1937, Meeting at New York, N. Y. ; received October 
1, 1937. 

** General Radio Company, Cambridge, Mass. 



response over the important frequency range, practically free from directional 
effects. Such a device is unaffected by ordinary changes of temperature and hu- 
midity, and even unusually low temperatures produce only a small change of sensi- 
tivity, for which correction can easily be made, if desirable. 

In the interests of convenience the microphone is mounted upon a folding 
bracket on the top of the sound-level meter, and turns down into a compartment 
cast in the panel when not in use. This makes it unnecessary to unwind any 
cables or plug in the microphone each time a measurement is made, since, under 
normal conditions of use, the microphone is always connected directly to the 
instrument. Provision is made, however, for removing the microphone from the 

FIG. 1. Type 7 59- A sound-level meter. 

sound-level meter and using it on a cable, for the few applications where such an 
arrangement is desirable. A special cable and tripod are available for this pur- 

The amplifier circuit itself is of the resistance-capacitance coupled type, using 
pentodes. By proper design of the screen-supply circuits an unusually high 
degree of stability of gain has been achieved, so that it is seldom necessary to reset 
the calibration control as the batteries wear out. Another feature of the amplifier 
circuit is the extremely low battery drain, the total plate current being only 2 
milliamperes. Naturally, this allows the use of very small batteries. Also, an 
amplifier circuit of this type requires no transformers or other heavy components. 

The use of a ballast tube in the filament circuit and of the stabilized amplifier 


circuit makes all filament current or battery adjustments quite unnecessary. 
This is a great convenience, since it is not necessary to adjust any battery controls 
when putting the instrument into operation. Push-buttons are provided on the 
panel for indicating directly on the "Decibels" meter the condition of the bat- 
teries a red line on the meter indicates when the batteries should be replaced. 
Accordingly, except for occasionally pushing the battery-test buttons to be sure 
that the voltages are sufficiently high to provide stable operation, it is never 
necessary to pay any attention to the batteries. When it does become necessary 
to change the batteries, the change is accomplished easily and quickly, since the 
battery-box cover is provided with spring contacts that automatically make the 
proper connections to the batteries when the cover is fitted in place. 

The use of small batteries and an amplifier circuit requiring no transformers 
results in an unusually light instrument. The elimination of transformers also 
minimizes the possibility of inductive interference. The main panel and micro- 
phone housing are a single aluminum casting which is light but rigid. The whole 
instrument is housed in an "airplane-luggage" case made of light-weight plywood 
covered with leatherette and completely shielded electrostatically. The resulting 
instrument is small and light, and attractive in appearance. 

Because of the high gain of the amplifier (approximately 140 db.) it is naturally 
necessary to make some provision for insulating the vacuum-tubes against me- 
chanical shock in order to minimize microphonic pick-up in the tubes themselves. 
This is accomplished by special rubber bushings which support the complete 
amplifier assembly. All the heavier parts of the amplifier, such as by-pass con- 
densers, etc., have been mounted on this assembly, and the rubber mountings have 
an extremely low deflection rate for small deflections, thus providing a long 
natural period of oscillation for the sub-assembly. The rubber mountings are so 
designed, however, that the deflection rate increases rapidly as the deflection is 
increased, thus providing a snubbing action and making it impossible to damage 
the instrument in ordinary handling or jarring. 

Convenience of operation has been considered of paramount importance in the 
design of the new sound-level meter. The main attenuator is adjustable in 10-db. 
steps, and, as previously mentioned, the indicating meter, which actually covers 
a range of 16 db., has a scale that is practically linear, which is achieved by the 
use of shaped pole-pieces. Three weighting networks are provided, in accordance 
with the A.S.A. standards, including the 40-db., 70-db., and flat networks. The 
first two of these networks are used for low-level and medium-level measurements, 
respectively, while the flat network is used generally for high-level measurements 
and for measuring frequency response. Fig. 1 shows the general appearance of 
the instrument. In particular, the arrangement of the panel, the clear lettering 
on the panel and on the meter scale, and the novel microphone mounting should 
be noted. All necessary operating instructions are fastened permanently inside 
the lid of the cabinet. 

One extremely convenient feature of the new instrument is the method of reset- 
ting the calibration. Although the amplifier used in this device has an unusual 
degree of stability, it was not considered desirable to rely upon this factor entirely 
for maintaining the permanence of calibration. Accordingly, provision is made 
so that the amplifier gain may be reset quickly and easily at any time. The 
arrangement consists essentially of applying a voltage through an attenuator to 

April, 1938] 



the input of the amplifier. The magnitude of this voltage and the output of the 
amplifier are indicated alternately on the "Decibels" meter when the calibration 
button is pressed. If the readings are alike the gain is correct. If the readings 
are not alike a screw-driver control may be adjusted to reset the calibration to its 
factory value. Ordinarily the alternating voltage for making this test is obtained 
from commercial power lines, a connecting cord being provided for the purpose. 
In the absence of a-c. power lines, however, an audio-frequency oscillator or other 
similar device may be used. 

In order to test the sound-level meter under actual field conditions, a large 
number of measurements have been made, including most of the important prob- 
lems to which the sound-level meter is applied (Fig. 2). The following informa- 
tion is not presented as an exhaustive survey of theater conditions, but merely to 
give an idea of conditions in what seems to be a typical suburban theater. The 
measurements were made in the University Theater at Cambridge, Mass., which 
has a seating capacity of about 2000 persons. 

to 100 1000 10,000 


FIG. 2. Variations of frequency response in center seats of motion picture 


The initial noise-level in this theater that is, with all air-conditioning and 
mechanical equipment shut off is about 26 db. A loud street noise, such as a 
pneumatic drill, may raise the level to as high as 32 db., which, as will be noted 
later, is considerably below the normal noise-level within the theater when the air- 
conditioning equipment is in operation. Accordingly, it appears that the theater 
is quite satisfactorily insulated from outside noises. 

Turning on the air-conditioning equipment raises the noise-level in the theater 
by nearly 20 db. in some locations. For instance, in the front orchestra seats the 
total noise-level with the air-conditioning equipment in operation is approxi- 
mately 45 db. The level, however, decreases toward the back of the theater, 
reaching a minimum of approximately 37 db. in the rear rows and 41 db. in the 


Measurements made of the loudness of reproduced speech and music in this 
theater seem to be about average. 2 Speech sounds, for instance, were around 70 
to 72 db., while music reaches 80 db. or higher. The background noise in the 
theater during the program, which includes, aside from the noises previously men- 
tioned, the audience noise and the noise from the reproducing system, is approxi- 
mately 54 db. 

Frequency characteristics were taken at various positions in the theater in order 
to determine the changes hi frequency response with location. Some of these 
data are shown in Fig. 2, which indicates clearly how the frequency response changes 
in the center of the theater between the front seats and the rear seats. The effect 
of the balcony upon the high frequencies is particularly noticeable and readily 
accounts for the decrease in articulation under the balcony. 

Data such as these are invaluable to the theater owner or operator, since they 
show readily how well the various portions of the audience are actually hearing 
the reproduced sounds. As a result of such measurements it is frequently possible, 
by proper acoustical treatment or by changes or additions in the speakers or 
tweeters, to improve noticeably the quality of reproduction throughout the 

The data shown here were obtained by merely connecting a beat-frequency os- 
cillator to the input of the amplifying equipment. Obviously, similar runs may 
be made when using a constant-frequency film in the projectors, thus obtaining 
an overall measure of the reproduction, including the optical equipment. No 
particular difficulties were encountered due to standing waves when making mea- 
surements at the higher frequencies, but there was some trouble from this source 
at the lower frequencies. Accordingly, it would be desirable, where extreme 
accuracy was warranted, to use a warble tone to minimize the effects of standing 

The authors wish to express their appreciation to Messrs. S. Sumner and C. W. 
Parshley for the use of the University Theater during these tests and to Mr. O. B. 
Asten of Electrical Research Products, Inc., for his cooperation in carrying out 
the tests. 


1 "The Technique of Noise Measurement," Bulletin 20, General Radio Co., 
Cambridge, Mass. 

2 WOLF, S. K., AND SETTE, W. J. : "Factors Governing Power Capacity of Sound 
Reproducing Equipment in Theaters," J. Soc. Mot. Pict. Eng., XV (Oct., 1930), 
No. 4, p. 415. 

WOLF, S. K., AND SETTE, W. J. : "Progress hi the Acoustics of Sound Recording 
and Reproduction for Motion Pictures," Rev. Sci. Instr., VII (Sept., 1936), No. 9, 
p. 323. Note that the decibel ratings mentioned hi these papers are referred to 
the threshold of audibility. The new sound-level meters use 10 ~ 18 watts per sq. 
cm. as reference level. Accordingly, for any given sound a new standard sound- 
level meter will read approximately 7 db. higher than if the measurement were 
referred to the threshold of audibility. 


MR. KELLOGG: To what level are the decibels referred; to the threshold of 
hearing or some arbitrary level. 


MR. SCOTT: The American Standards Association has, within the past year 
or so, set a new standard reference level, which is about 7 db. below what we used 
to regard as the threshold. The new level is 10 ~ 16 watt per sq. cm. Compared to 
earlier sound-level meters, this one reads about 7 db. higher. 

MR. RICHARDSON: What are supposed to be the practical benefits of the in- 
strument in a theater? Does it in any way aid the projectionist in adjusting 
the sound-level? 

MR. SCOTT: In some theaters the instruments have been found quite useful 
for checking the level of reproduction as well as for making other measurements 
as described in the paper. It is possible to mount the microphone anywhere in 
the auditorium and run an extension cable to the instrument, which can be in 
the projection room, and thus to keep an actual acoustical check on the level 
in the auditorium. 



The change-over system described here has been developed for the purpose of 
eliminating faulty change-overs in the process of projecting motion picture film. 
This equipment makes instantaneous, precise, automatic change-overs from one 
projector to another, and never fails to make the complete change-over at exactly 
the correct moment. 

The system eliminates necessity for visual cue marks of any type, such as are 
now placed on film by projectionists and by producers in processing the film, 
and consequent replacements by the exchanges due to such marking and mutila- 
tion are avoided. The system also reduces fire hazard considerably. 

Change-overs done by this system are not visible to the eye. Faulty changes 
from one projector to the other are entirely eliminated. Not a single frame of 
the picture or one spoken word of the sound record is lost. Fade-out changes are 
properly timed, and perfect continuity of both picture and sound is assured. 

More efficient use of curtains, lights, and other effects used in modern motion 
picture theaters is afforded by relieving the projectionist of much eye-strain and 
tension in his work and permitting him to devote more attention to the actual 
management of his projection room and to the supervision and maintenance of 
his equipment. 

Projectionists have often been unjustly criticised for their inability to see the 
indistinct and mutilated cues placed upon the films. Under the best of condi- 
tions, cues can be missed. The cues inserted by the producers are often indistinct 
because of the background. Many films, therefore, are mutilated by the pro- 
jectionists in their many efforts to create satisfactory cue marks. These cues 
may consist of grease pencil marks, sticker, punches, scratches with sharp in- 
struments, notches cut out of the edges, or tinfoil glued to the film. They may 

* Presented at the Spring, 1937, Meeting at Hollywood, Calif. 
** Automatic Change-Over Company, Los Angeles, Calif. 


be placed at various distances from the end of the film, and the last projectionist 
running such film is left with the alternative of calculating and interpreting the 
previous timing, or of adding a new mutilation or cue-mark of his own. 

The equipment is fully automatic in operation. It is not necessary for the 
projectionist to operate any switch, or otherwise assist in the functions of starting 
the projector motors, or of timing and operating the dowsers and fader in their 
proper order. The electrical circuit is so arranged that the regular motor switches, 
dowser switches, and sound change-over device (fader or key) may be used at 
any time independently of the A. C. O. Upon installation, the regular controls 
usually employed in the projection room need not be changed, removed, or altered 
in any way. 

The equipment consists of the following units which combine to render the 
device completely automatic: 

FIG. 1. Mounted film cue clips, 
finger up. 

Film cue-clips are attached by means of a side-plate to the film reels. When 
set by the projectionist when the film is rewound, these clips predetermine the 
exact time at which the desired change-over shall take place. The unwinding 
film releases each finger or cue-clip at the proper time. The release of the finger or 
cue-clip actuates plunger shafts within the spindle and operates mercury switches 
contained within the top-magazine switch housing. The mechanical features of 
this arrangement have been thoroughly tested over a period of several years and 
have proved to be dependable and satisfactory. 

The system includes a pair of top-magazine spindle assemblies and mercury 
switch housings, one of which is installed on each projector in place of the regular 
spindle on the top-magazine bracket. The assembly is easily mounted on the 
top-magazine hanger casting with two set-screws. The A. C. O. spindle takes 
the place of the old one. No electric wires, contactor, switches, etc., are used 
inside the upper film magazine. The spindle shafts are standard for use on any 
35-mm. projection equipment. Spindles are fitted with a specially designed 


brake mechanism, permitting exact adjustment of film tension at any time even 
while the projector is in operation. 

Economies that the A. C. O. can be expected to effect are noticeable in the 
saving of carbons and electric power. No long period of waiting is required 
for visual motor-start and change-over cues to appear upon the screen, during 
which time the operator must necessarily burn both projector arcs. 

The A. C. O. System performs three distinct operations, automatically: 
(1) At a proper predetermined point the A. C. O. takes its cue from the film 
being projected, and, automatically actuates a switch, starting the oncoming 
projector motor. This motor-start may indicate to the projectionist that he 
should strike the arc on the incoming projector. 

FIG. 2. Top-magazine spindle (open) mounted 
on reel. 

(2) At the next predetermined point the A. C. O. takes another cue from the 
film being projected, and actuates the switch that controls the dowsers. 

(5) At the second predetermined point the switch controlling the sound 
change-over is actuated. Both the dowsers and the sound change together. 

There can be neither black nor white screens, nor the loss of a single spoken 
word. Being electrical and mechanical in operation, change-overs made by the 
A. C. O. are not subject to human failure or error and leave nothing to chance. 

The control cabinet unit (115 volt a-c. 3 amp.) is mounted on the projection 
room wall, usually between the projection machines, conveniently accessible to 
the projectionist. It comprises a centralized control, through automatic me- 
chanical-electrical interlocks, which governs the operations of starting and 
stopping the motors and making the various changes of dowsers and sound. A 
convenient switchboard enables the projectionist to operate the equipment at 
will with the greatest flexibility and selectivity. 



The dowsers are designed to fit each projector head and are a standard part 
of the equipment. They are small and compact, neatly designed, durably con- 
structed, and very efficient in operation. The dowser shutter travels only 3 / 4 

FIG. 3. Main control cabinet ; (left) exterior; (right) interior. 

inch and its action is very fast and practically noiseless. The magnetic coils 
have a large overload capacity and all fittings are extra-heavy. The dowsers 
are specially designed to operate with the A. C. O. change-over equipment, but 
may also be operated separately and independently if desired. Any other 

FIG. 4. A. C. O. dowser. 

standard dowser will also operate satisfactorily when used in conjunction with 
A. C. O. equipment. 

The A. C. O. equipment has been thoroughly tested in actual theater operation 
over a period of six years, and has proved conclusively that it meets all of the 
essential requirements. 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C. 

British Journal of Photography 

85 (Jan. 14, 1938), No. 4054 
Reducing Projector Noise (16-mm.) (p. 25). 


18 (Jan., 1938), No. 1 

Design of Resistance- Coupled Amplifiers (p. 11). 
Amplification Problems of Television (p. 15). 
Production Development of Television Tubes (p. 24). 


11 (Jan., 1938), No. 1 
Reviewing the Video Art (p. 8). 
Wide-Band Television Amplifiers (p. 16). 
Cathode-Ray Phasemeter (p. 24). 

Home Facsimile Recording (p. 26). 









13 (Dec., 1937), No. 18 
Aufgaben des Kinoverstarkers (Motion Picture Amplifier 

Problems) (p. 205-6) R. WIGAND 

International Projectionist 

13 (Jan., 1938), No. 1. 
Analyses of Modern Theater Sound Reproducing Units 

(p. 14). A. NADELL 

Film Aperture Decision Rests 111 with Rest of Experts 

(p. 20). H. GRIFFIN 

Projecting Hi- and Low-Range Prints; Standard Fader 
Setting Data (p. 24). 

Journal of the Acoustical Society of America 

9 (Jan., 1938), No. 3 
An Improved Magnetostriction Oscillator (p. 185). 



468 CURRENT LITERATURE [j. s. M. p. E. 

Theoretical and Experimental Investigation of the Trans- 
mission of Sound over Reflecting Surfaces (p. 193). G. W. PIERCE AND 

A. No YES, JR. 
Variations in Sound Absorption Coefficients as Obtained 

by the Reverberation Chamber Method (p. 234). R. M. MORRIS, 

The Predetermination of Speech Levels in Auditoria with 

Coupled Spaces (p. 244). G. E. MORISON 

Journal of the British Kinematograph Society 

1 (Dec., 1937), No. 1 
The Design of Special Telephoto and Wide Aperture Lenses 

(p. 3). K. J. HABELL 

The Functions of a Director in the Making of a Film (p. 8). V. SAVILLE 
The Standardization of Camera Exposure in Kinematog- 

raphy (p. 20). P. E. SMETHURST 

Studio Control Systems (p. 33). M. F. COOPER 

Kinematograph Weekly 

251 (Jan. 13, 1938), No. 1604 

All Facilities Offered by the Denham Laboratories (p. 163). 
Beauty of Negative-Positive Dufaycolor (p. 165). 

Kinematograph Weekly 

251 (Jan. 20, 1938), No. 1605 

A Survey of Last Year's Technical Progress (p. 63). R. H. CRICKS 

New Camera and Disk Recorder (p. 68). 


19 (Dec., 1937), No. 13 

Zur Theorie des Donnereffektes und seiner Abhangigkeit 
vom Gammawert (Theory of the Donner Effect and Its 
Dependence on the Gamma Value) (p. 305). A. NARATH 

Einfluss der photographischen Behandlung auf den Don- 
ner-Effekt von Tonaufnahmen in Zackenschrift (Influ- 
ence of Photographic Treatment on the Donner Effect 
on Sound Exposures in Variable- Width Recording (p. 



Motion Picture Herald (Better Theatres Section) 

130 (Jan. 8, 1938), No. 2 

Relative Characteristics of Low- and High-Intensity Pro- 
jection Light (p. 30). E. R. GEIB 

130 (Feb. 5, 1938), No. 6 

The Practical Theater Man's Approach to Acoustic Prob- 
lems (p. 23). 

A New Maintenance Weapon; Electrostatic Air Cleansing 
(P- 30). 

April, 1938] CURRENT LITERATURE 469 

What to Do While Waiting for the Sound Service Engi- 
neer (p. 35). A. NADELL 

Photographische Industrie 

36 (Jan. 12, 1938), No. 2 

Riickblick auf die Kinotechnik 1937 (Glance at Motion 
Picture Technical Progress for 1937) (p. 45). 

RCA Review 

2 (Jan., 1938), No. 3. 

Direct-Viewing Type Cathode-Ray Tube for Large Tele- 
vision Images (p. 289). I. G. MALOFF 
Television Cathode-Ray Tubes for the Amateur (p. 297). R. S. BURNAP 
A New System of Remote Control (p. 303). C. N. KIMBALL 
Effects of Space-Charge in the Grid-Anode Region of 
Vacuum Tubes (p. 336). B. SALZBERG AND 



11 (Jan., 1938), No. 119 
A New Emitron Camera with Greatly Increased Sensitivity 

(P- ID- 

Projection Tubes (p. 13). V. K. ZWORYKIN 

AND W. A. 

How Cossor Television Receivers are Tested (p. 24). 
ABC of Magnetic Scanning (p. 26). G. PARR 




Officers and Committees in Charge 

W. C. KUNZMANN, Convention Vice-President 

J.I. CRABTREE, Editorial Vice-President 

G. E. MATTHEWS, Chairman, Papers Committee 

W. WHITMORE, Chairman, Publicity Committee 

E. R. GEIB, Chairman, Membership Committee 

N. D. GOLDEN, Chairman, Local Arrangements Committee 

Local Arrangements and Reception Committee 


N. D. GOLDEN, Chairman 





Registration and Information 

W. C. KUNZMANN, Chairman 



Ladies' Reception Committee 

MRS. R. EVANS, Hostess 

assisted by 




Banquet Committee 

R. EVANS, Chairman 



Publicity Committee 

W. WHITMORE, Chairman 




Convention Projection Committee 

H. GRIFFIN, Chairman 




Officers and Members of Washington Projectionist Local 224. 

Apparatus Exhibit 

S. HARRIS, Chairman 

Membership Committee 

E. R. GEIB, Chairman 


Hotel and Transportation Committee 

J. G. BRADLEY, Chairman 




The headquarters of the Convention will be the Wardman Park Hotel, where 
excellent accommodations are assured. A reception suite will be provided for the 
ladies, for whom also is to be arranged an interesting program of entertainment. 

By special arrangement with the Hotel Management, special breakfast, lunch- 
eon, and dinner service will be provided on the Continental Room Terrace, 
for SMPE delegates only. 

The following daily hotel rates, European plan, are guaranteed to SMPE 
delegates attending the Convention: 

One person, room and bath $ 3 . 50 

Two persons, standard bed 5.00 

Two persons, twin beds 6.00 

Parlor suite, one person 9 . 00 

Parlor suite, two persons 11 . 00 

472 SPRING CONVENTION [j. s. M. P. E. 

Room reservation cards will be mailed to the membership of the Society in the 
near future, and those who plan to attend the Spring Convention should return 
their cards promptly to the Wardman Park Hotel to be assured satisfactory 
accommodations. Local railroad ticket agents should be consulted with regard to 
trains and rates. 

For those who will motor to the Convention ample free parking space is avail- 
able on the Hotel grounds. For those who prefer parking in the Hotel garage 
a special rate of 75 cents a day has been arranged. 

Technical Sessions 

An attractive and interesting program of technical papers is being assembled 
by the Papers Committee. All technical sessions, apparatus symposiums, and 
film programs will be held in the Little Theatre of the Hotel. 

Apparatus Exhibit 

An exhibit of newly developed motion picture apparatus will be held, to which 
all manufacturers of equipment are invited to contribute. No charge will be 
made for space. Information concerning the exhibit and reservations for space 
should be made by writing to the General Office of the Society. 

Apparatus displayed should be newly designed or developed, or should have 
features of technical interest for the engineers attending the Convention. 

Registration and Information 

The Convention registration headquarters will be located at the entrance of the 
Little Theatre, where all the technical sessions will be held. The members of the 
Society and guests attending the Convention are expected to register and receive 
their badges and identification cards for admittance to special evening sessions. 
These cards will also be honored at several de luxe motion picture theaters in 
Washington during the four days of the Convention. 

Informal Luncheon and Semi-Annual Banquet 

The usual informal Luncheon will be held at noon of the opening day of the 
Convention, April 25th, in the Continental Room of the Hotel. On the evening 
of Wednesday, April 27th, will be held the Semi-Annual Banquet of the Society, 
also in the Continental Room, at 8:00 P.M. Addresses will be delivered by 
prominent members of the industry, followed by dancing and other entertain- 

April, 1938] SPRING CONVENTION 473 

Motion Pictures 

Delegates registering at the Convention will be supplied with complimentary 
passes to the following motion picture theaters in Washington during the dates of 
the Convention: 

By courtesy of Mr. J. J. Payette: Warners' Uptown and Earle Theaters. 

By courtesy of Mr. H. Meiken: RKO Keith's Theater. 

By courtesy of Mr. C. Barron: Loew's Capitol, Palace, and Columbia Theaters. 

Ladies' Committee 

A number of interesting events are being planned by Mrs. R. Evans, Hostess, 
and the Ladies' Committee. On Monday, April 25th, at 5 P.M. Mrs. Franklin 
D. Roosevelt has kindly consented to receive the ladies of the Convention at the 
White House. All those who intend to be present at the reception should trans- 
mit their names as early as possible to Mr. W. C. Kunzmann, Convention Vice- 
President, at the General Office of the Society, Hotel Pennsylvania, New York, 
N. Y. 


The Wardman Park Hotel management is arranging for golfing privileges for 
SMPE delegates at several courses in the neighborhood. Regulation tennis 
courts are located upon the Hotel property, and riding stables are within a short 
distance of the Hotel. Trips may be arranged to the many points of interest in 
and about Washington. 

Points of Interest 

To list all the points of interest in and about Washington would require too 
much space, but among them may be mentioned the various governmental 
buildings, such as the Capitol, the White House* Library of Congress, Depart- 
ment of Commerce, U. S. Treasury, U. S. Bureau of Standards, Department of 
Justice, Archives Building; and other institutions such as the National Academy 
of Sciences, the Smithsonian Institution, George Washington University, Wash- 
ington Cathedral, Georgetown University, etc. In addition may be included the 
Lincoln Memorial, the Washington Monument, Rock Creek Park, The Francis 
Scott Key Memorial Bridge, Arlington Memorial Bridge, the Potomac River, 
and Tidal Basin. Mt. Vernon, birthplace of Washington, is but a short distance 
away and many other side trips may be made conveniently via the many highways 
radiating from Washington. 



APRIL 25-28, 1938 

The Papers Committee submits the following abstracts of papers for the considera- 
tion of the membership. It is hoped that the publication of these abstracts will en- 
courage attendance at the meeting and facilitate better discussion of the papers. 


G. E. MATTHEWS, Chairman 
L. A. AICHOLTZ, Chairman, West Coast 







Report of the Progress Committee; J. G. Frayne, Chairman. 

The outstanding event in cinematography during the past year was the devel- 
opment of the high-speed panchromatic emulsions by the Agfa Ansco Corporation. 
Other interesting advances in the emulsion field are the development of two fine- 
grain duplicating film stocks by the Eastman Kodak Company. Of interest also 
is the new sound emulsion developed by Dupont in which the periodic variation 
in sensitivity brought about by the present emulsion-drying methods has been 

In the sound -recording field, items of interest are the introduction of linear deci- 
bel volume indicators by United Artists Studio and the introduction by RCA of 
the modulated high-frequency method of determining optimal processing condi- 

" Sound Stages and Their Relation to Air-Conditioning"; C. M. Wert and L. L. 
Lewis, Carrier Corp., Syracuse, N. Y. 

The development and growth of the modern motion picture sound stage has al- 
most paralleled that of sound pictures. Weather and advancement of lighting 
technic undoubtedly brought about original need of enclosed stages. Advent of 
sound recording brought about requirements not originally considered. Modern 
sound stages have increased not only in quality but in size. The modern sound 


stage must have structural strength to withstand the elements, including earth- 
quakes. It must meet requirements of set construction, sound-proofing, and 
occupancy. Sound treatment makes necessary other treatment for satisfactory 
occupancy. Lighting on the sound stage is the greatest contributor of heat gain 
within the stage. Lighting is variable as to amount and duration, and must be 
controlled correctly. Size and number of sets are very variable and create their 
individual problems. Both the number and types of persons present on a sound 
stage play their parts in the relation between air-conditioning and the sound 

Construction that retards the flow of heat in either direction through walls 
necessitates the addition and removal of the heat. Lighting on a sound stage is of 
such magnitude that its effects must be removed. High-salaried personnel, often 
in costume, demand comfort while working. Management is obviously economi- 
cally in better position if personnel is comfortable ; less time is lost due to make- 
up retouching and less delay brought about by perspiration dampened costumes. 

An air-conditioning system should have the ability to provide heating, cooling, 
ventilation, and cleaning. Heat in the air rising to the top of the stages should be 
removed by an exhaust system. Stages are generally maintained at 75 F and 
50 per cent relative humidity, with temperature settings above and below at the 
option of the occupants. Floor distribution of air has the advantage of more 
economical removal of rising heat but has the practical disadvantage of placing 
set construction and personnel too near source of cooling. Overhead distribution 
has the advantage of better temperature distribution but is less economical in the 
removal of rising heat from lights. 

Sound treatment of an air-conditioning installation is necessary for continuous 
operation of the system. If the system does not operate continuously the heat 
load builds up to the point where the system can not adequately regain comfort- 
able conditions during non-shooting periods. Treatment is accomplished by 
both isolation and absorption of generated sound, and can be so accurately deter- 
mined that a guarantee of the increase in noise level can be given in decibels and 
in relation to frequency ranges. 

"Motion Picture Projection from Metallic Film"; R. W. Carter, Taylor- 
Sloane Corp., New York, N. Y. 

A brief history is given of the various processes for putting photographic images 
on metallic surfaces and the evolution from flat surfaces to flexible metal ribbons 
is discussed. The subject of metal films is traced under the following headings: 
The physical and mechanical difficulties in the development of a metal strip 
suitable for projection. The physical, chemical, and mechanical properties neces- 
sary for the photographic emulsions and photographic developers. The ef- 
fect of mechanical strain and the heat of the projection machine upon the metal 
film. The relative wearing quality* of metal film as compared with that of cellu- 
lose film. The possibilities of coating both sides of the metal strip and the de- 
velopment of printing machines to print thereon. Making original master 
negatives on standard photographing equipment. Dubbing positive prints from 
the master metal negative. The optical system best adapted for getting the high- 
set possible reflection from the polished surface of the metal film. The comparison 
of light transmission from celluloid and metal films. The effect of heat upon the 

476 SPRING CONVENTION [j. s. M. p. E. 

image on a metal film. Can a metal film be joined rapidly if it conies apart? A 
comparison of shrinkage between metal film and cellulose film. What evidence 
have we of the permanence of metal film? Will it be possible to develop color on 
metal film, and will the use of prisms make it possible for successful projection? 
What changes will the operator have to make in technic and general prac- 
tice? Why will the sound be more accurate from a reflected image? Will it be 
possible in the future to use a series of sound-tracks in various languages on the 
metal film? With the elimination of the fire hazard, shrinkage, and the introduc- 
tion of less weight and positive permanence, what are the chief defects to be 
expected in metal film, and what is proposed to overcome these defects? 

"Documentary Film Study a Supplementary Aid to Public Relations"; A. A. 

Mercey, School of Public Affairs, The American University, Washington, D. C. 

Documentary films are proving of increased importance as a factor for inform- 
ing and mobilizing opinion. The marked success of two U. S. documentary films, 
The Plow That Broke the Plains and The River, both written and directed by Pare 
Lorentz, has focused new attention upon this type of film. The school of Public 
Affairs of American University conducts an "in-service" training school for gov- 
ernment employees whereby registrants obtain instruction in courses and subjects 
from experts in various Federal departments. Included in these curricula are a 
series of courses on public relations. The film as a factor in public relations is an 
important one. In answer to requests for some information and instruction in this 
new field, a course in "Documentary Films Today" was instituted. 

The film course included an eight- week study with screenings, film analyses, and 
discussions conducted by visiting experts in film -making and film use. The sub- 
jects covered were : The newsreel as contemporary historian ; the March of Time 
as a document; federal, educational, and scientific films; U. S. Government docu- 
mentary films; documentary aspects of Hollywood films; foreign documentaries; 
industrial, sales, and domestic propaganda films. During the eight-week period, 
visiting experts included a government producer, an industrial film user, an educa- 
tor, and others. Technical aspects with reference to advances in film production 
were discussed. 

In addition to regular film discussion and study, a number of reports were made 
on documentary film activities. Among the most important was one on a federal 
film survey. For the first time, a complete survey of all U. S. government films 
is being made that will compile in one place the data on motion pictures. A 
standardized type of procedure was outlined. 

"The Determination of Correct Exposure in Photography"; L. A. Jones, Kodak 
Research Laboratories, Rochester, N. Y. 

Many treatments of this subject, some dealing with certain specific phases, and 
some fairly complete, are to be found in varidus textbooks and scientific journals 
in the field of optics and photography. In spite of this, however, there seems to be 
some uncertainty in the minds of some relative to the correct manner of dealing 
with the problem. The present treatment is distinctly of a tutorial character, an 
endeavor being made to present the problem in a clear and systematic fashion. 
Much of the existing confusion is doubtless due to the multiplicity of photometric 
units found in the literature of photometry, and to a certain amount of ambiguity 

April, 1938] SPRING CONVENTION 477 

in the current definitions relating to these units. An attempt is made to present 
a considerably simplified conception of the minimum number of photometric 
quantities required for dealing with the exposure problem. The relation between 
image illumination and object brightness is dependent upon several physical char- 
acteristics of the image-forming system. Quantitative information relating to 
specific image-forming systems and a general average image-forming system useful 
for computing the relation between object brightness and image illumination are 
given. The relation of the sensitivity of photographic materials to the problem 
is considered in some detail, as well as the photometric and contrast characteristics 
of various types of photographic subjects. 

"Latent-Image Theory and Its Application to Low-Intensity Photographic 
Exposures"; W. J. Albersheim, Electrical Research Products, Inc., New York, 
N. Y. 

In a previous paper by the writer, it was shown that the photographic exposure 
characteristics are in agreement with the assumption that a photographic grain 
must absorb two photons of visible light in order to become developable. In the 
present paper, this theory is compared with recent physical research by other 
authors. It is assumed that a film grain is "sensitized" by the first absorbed 
photon and fully "exposed" by the second absorbed photon. 

Reciprocity -law failure at low-intensity exposures can be explained by the as- 
sumption that the sensitized state of film grains is unstable and that the number of 
sensitized grains decreases with time in an exponential manner unless fixed by 
activation. The half-time of this fading for certain emulsions is deduced from 
Kodak publications on reciprocity-failure characteristics. 

Conclusions from this theory are drawn with regard to the contrast improve- 
ment for low-intensity photography, such as astronomical work or newsreel photog- 
raphy under unsatisfactory lighting conditions, by pre- or post-fogging. The 
theoretical conclusions are checked with test results. 

"Effect of Aeration on the Photographic Properties of Developers"; J. I. Crab- 
tree and C. H. Schwingel, Eastman Kodak Co., Rochester, New York. 

Unseasoned elon-hydroquinone developers of relatively high alkalinity (pU. 
10.0 to 10.5) showed a rapid decrease in activity after aeration for l l /z hours while 
elon-hydroquinone-borax developers of low alkalinity (pH 8.4 to 8.8) showed in- 
creased activity (due to the liberation of alkali resulting from oxidation) which 
then remained constant for prolonged periods. 

In general, the alkalinity of developers containing hydroquinone increased on 
aeration, while those containing only elon showed little change. 

Practical tests with processing machines equipped with air agitation devices 
have shown that very constant developing conditions can be maintained with 
both positive and negative types of developers. 

"Solution Agitation by Means of Compressed Air"; C. E. Ives and C. J. Kunz, 

Eastman Kodak Co., Rochester, N. Y. 

In the development of motion picture film, the developer in the emulsion under- 
goes exhaustion and thereby loses activity. Agitation of the developing solution 
in the vicinity of the film is required to assure sufficiently rapid and uniform re- 

478 SPRING CONVENTION [j. s. M. P. E. 

newal by relatively fresh developer brought from the remainder of the bath. 

The present work is concerned with a method of bringing about this agita- 
tion by means of compressed air which is released at one or more points in the de- 
veloper, through which it rises to the upper surface creating a generally turbulent 
condition and setting up rapid streaming effects. 

The effectiveness of the stirring is limited by the tendency of the induced 
stream to form a narrow channel in one portion of the tank with relatively low 
velocity in the remainder of the tank. 

Various means have been tried in an effort to direct the rapidly moving stream 
along the film surface, and this was accomplished by means of a gridwork of con- 
ducting pipes extending from top to bottom of the rack and parallel to 
the sides of the racks. Tests for uniformity of development made by means of 
uniformly flashed film showed the benefit conferred by the various improvements 
in control of the agitation. Dimensions and details of construction are given for 
making up the distributing grid. 

"Maintenance of a Developer by Continuous Replenishment"; R. M. Evans, 
Kodak Research Laboratories, Rochester, N. Y. 

By a series of simple assumptions that do not depart appreciably from current 
practice it is shown that the concentration of any ingredient in a developer solu- 
tion that is continuously replenished during use may readily be calculated. The 
equations for the equilibria and rates of growth of the various substances are de- 
rived and application is made to a practical case. The benefits of chemical analy- 
ses for developer constituents both for maintenance of quality and for economy 
are pointed out, and the analytical methods published by Lehmann and Tausch 
are briefly outlined. 

"The Effect of />H upon the Washing of Processed Films"; S. E. Sheppard and 
R. C. Houck, Kodak Research Laboratories, Rochester, N. Y. 

Advantages stated to be obtained by adjusting fixing baths and wash-water to 
the isoelectric point of gelatin have been claimed. The advantages are said to be 
shorter washing time, less swelling and retention of water, with consequent im- 
provement in the jelly strength of the wet emulsion, and reduced drying time. In 
the present investigation the conditions as to pH of the solutions, and wash-water, 
rate of flow of water, residual thiosulfate, etc., were controlled accurately. The re- 
sults indicate that with a regular acid fixing and hardening bath (F-25) there is no 
advantage, but rather a disadvantage in washing at the isoelectric point (ca H 
4.9) rather than at H 7 to 8, since the time required to remove hypo to the same 
degree is increased, nor is less water retained. In a non-hardening acid fixing bath, 
there was little difference in washing time, but some gain in drying time for the 
isoelectric wash because of reduced water absorption. 

"A New Densitometer"; H. Neumann, Klangfilm G. m. b. H., Germany. 

Density measurements of variable-width sound records should cover a large 
range of densities, and the measuring area should be as small as possible, so as to 
make it easy to find a suitable area on normal sound records. 

The densitometer described, which is intended mainly for use in studios and 
laboratories but which is so accurate that it may be used also for scientific 

April, 1938] SPRING CONVENTION 479 

research, is capable of measuring densities of 0.01 to 2.5 of areas 2.5 mm. long and 
0.03 mm. wide, limited by a mechanical slit. The absorption of light by the object 
is determined by means of the current set up in a blocking layer photoelectric cell 
which is measured by a very sensitive galvanometer giving direct density readings. 
The calibration of the light-source can be checked very simply by a separate light 
path without making necessary removal of the object during the check. A special 
arrangement is provided for visual observation of the measuring area under the 

The density values are determined with parallel light, and from these data the 
values for diffuse light may be easily calculated. 

"The Transmission of Motion Pictures over a Coaxial Cable"; H. E. Ives, 

Bell Telephone Laboratories, New York, N. Y. 

The transmission of television signals over wire lines a number of years ago used 
signals corresponding to images of coarse detail, and required frequency bands ac- 
commodated by existing types of circuits. The television images now considered 
necessary correspond to frequency bands of greatly increased width, and will re- 
quire special wire networks and transmission means. 

The coaxial conductor recently in operation for experimental purposes between 
New York and Philadelphia can transmit a band of frequencies of approximately 
1000 kc. While designed primarily for multiple telephone channels, it offered 
the possibility of transmitting a single wide band as required for television. 

The experiment consisted in providing television-type terminal apparatus for 
producing signals falling within the available band, and of developing and utilizing 
methods of transmission that would make most complete use of the frequency 
band available. For convenience in the experimental work, the signals were 
generated from motion picture film. The film was scanned mechanically by 
means of a lens disk containing 240 lenses. The film was moved continuously at 
24 frames per second, and its motion, together with the motion of the lenses in the 
disk, swept each frame of the film in 240 juxtaposed lines. Light passing through 
the film was received on a photosensitive surface ; the resulting photoelectric cur- 
rent was amplified and by means of modulating and demodulating apparatus 
transmitted as a single sideband lying between approximately 150 and 950 kc. 
At the receiving end the single sideband signal was restored as a signal from zero 
to 800 kc. 

For reception, special cathode-ray tubes were used in which particular atten- 
tion was paid to the definition of the spot and the linearity of response. Synchro- 
nism between the two ends was obtained by sending a single frequency over a 
separate channel and using it to operate sweep circuits at the receiving end. The 
use of mechanical scanning and the high-definition receiving tubes resulted in pic - 
tures of very satisfactory quality within the limitations set by the frequency band . 
(Illustrated with slides and motion pictures.) 

"The Inter-Relationship of the Various Aspects of Color"; L. A. Jones, Kodak 
Research Laboratories, Rochester, N. Y. 

An understanding of the subject of color and color measurement involves a 
knowledge of many and diverse phenomena. Pursuit of this knowledge leads into 
many fields of physical or objective science, such as physics, physiology, biology, 
chemistry, etc., as well as into the domain of a subjective science, psychology. 

480 SPRING CONVENTION [j. s. M. p. E. 

While it may not be possible or even desirable to attempt to draw sharp lines of 
demarcation between all the various aspects of the problem, it does seem desir- 
able, for the sake of orderliness and clear thinking, to suggest a certain division of 
the subject into a few definite categories and to attempt to define the relations that 
exist between the various aspects of the problem as a whole. 

The present treatment of the subject is designed largely as a means of estab- 
lishing orientations in the general field. An attempt is made to develop a logical 
and unambiguous nomenclature that will enable us to discuss various aspects of 
the subject without the confusion that exists so generally at the present time when 
individuals of diverse trainings and viewpoints attempt to discuss the subject of 
color. The subject-matter divides itself rather logically into three clear-cut 
categories, which may be referred to as the physical, psychophysical, and psy- 
chological. Attention is drawn to the relation existing between the correlated as- 
pects in each of these three categories. An attempt is made to clarify the purely 
physical factors involved and to discuss certain sensory and perceptual aspects of 
color and the relations existing between them and their physical and psychophysi- 
cal correlates. 

"The Theory of Color Reproduction"; A. C. Hardy, Massachusetts Institute oj 
Technology, Cambridge, Mass. 

All methods of three-color photography are the outgrowth of a suggestion made 
in 1855 by Clerk Maxwell, the illustrious British physicist. The method that he 
suggested would now be classed as an additive process, since the final reproduction 
was effected by projecting three lantern-slides in register on the same screen; one 
lantern being supplied with a red filter, one with a green filter, and one with a blue 
filter. Maxwell suggested further that these lantern-slides be prepared from three 
negatives, each negative being exposed through the same filter that was to be used 
in projecting the corresponding lantern-slide. An extension of Maxwell's reason- 
ing to subtractive processes leads to the conclusion that the dyes used in the pro- 
duction of the positive images should each be complementary in color to the cor- 
responding taking filter. 

Despite Maxwell's intimation that his process was theoretically incapable of 
perfect reproduction, the basic features of Maxwell's reasoning have been incor- 
porated into the commonly accepted theory of color reproduction. The recent 
progress in the science of colorimetry has made it possible to investigate the rela- 
tion that should obtain between the characteristics of the taking filters and the 
colors of the reproduction primaries. Such an investigation shows that the taking 
filters required for perfect reproduction have characteristics that are very different 
from those in common use. 

The paper is concerned with the establishment of the conditions that lead to 
faithful reproduction by any three-color process. Examples of the application of 
these fundamental conditions are given for both additive and subtractive processes. 

"Screen-Film Negative-Positive Process"; T. T. Baker, Dufaycolor, Inc., New 
York, N. Y. 

Progress in two directions has greatly simplified the making of prints from 
screen-film negatives. The study of emulsion characteristics and of the mechanics 
of development with silver bromide solvents has led to the avoidance of color di- 

April, 1938] SPRING CONVENTION 481 

lution in copying one screen material from another. Sodium thiosulfate in a 
metol developer has been shown to localize development in the lower stiata of the 
film, so that the silver image is formed in close contact with the reseau, largely 
eliminating scatter at the boundaries of differently colored units; the crystalline 
structure of the silver salts and grain-size frequency also assists in preventing scat- 
ter. Residual color dilution as the result of the 45-degree oriented reseaux is ex- 
plained, and the way in which this has been counteracted by suitable choice of 
gammas in the negative and positive material. The production of a vapor-lamp 
emitting the line spectra of mercury and cadmium without appreciable spectral 
background, combined with a liquid didymium chloride filter has provided a triple 
monochromatic light-source, the spectral lines of which coincide with the peaks of 
the reseau transmissions, thereby eliminating dilution of color due to overlap, such 
as has always previously been present with color filters of the narrow-cut type. 
The Dufaycolor contact printing machine with automatic control of both hue and 
printing light is described. The technics of printing, and development with 
standard equipment, are described with lantern-slides and projections of recent 
35-mm. cine prints (which are at present circulating in English theaters). 

"Problems Involved in Full-Color Reproduction of Growing Chick Embryo"; 
E. S. Phillips, New York State College of Agriculture, Cornell University, Ithaca, 
N. Y. 

Attempts to record on 16-mm. color film the physiological changes that take 
place during the 21-day incubation period of the hen's egg presents problems vary- 
ing with each day's growth. Because the author was working with living subjects 
that required strict adherence to narrow tolerances in order to maintain normal 
embryological development and even life itself, it was necessary to adapt photog- 
raphy to the problem. 

Development of the embryo is shown in three different ways, i. e., (1) trans- 
mitted light, with shell entire; (2} removal of part of the shell, and subsequent 
photography by reflected light; and (3) removal of the entire shell, and placing 
the embryo in a watch crystal, thus showing all parts in their relative sizes. 

In all three methods, temperature, humidity, and light control constitute the 
major problems. 

"The Multiplane Camera"; W. E. Garity, Walt Disney Productions, Ltd., 
Hollywood, Calif. 

In line with the policy of continued improvement in cartoon technic, it was rec- 
ognized that several developments could be undertaken which if successfully 
adapted, would add much to the power and charm of animated motion pictures. 
By confining cartoon photography to a single plane in front of the camera, the ex- 
pense and difficulty of creating a convincing illusion of depth and a real-life ap- 
pearance of camera movement made the consideration of a several-plane technic 
imperative. The out-of-focus diffusion and the differential movement of fore- 
ground and background elements of scenes can be attained most easily by sepa- 
rating those elements on different planes in front of the camera. The problem 
resolved itself into the adaptation of glass-shot technic to cartoon production. 
In separating the scene elements into several planes, many other advantages were 
gained, such as the lighting control of single-scene elements, the ease of using spe- 


cial-effects equipment, and the possibility of using backlight and process back- 

The answer to the problem was a permanent machine, the multiplane camera, 
which was built to use the standard cartoon technic of animated characters in con- 
nection with several plane backgrounds. The machine was built with the view of 
accuracy of control, complete flexibility of scene set-up, and efficiency of operation. 
This required plane elements that could be quickly and accurately assembled, an 
accurate indication system, and an interlocked system of control. 

Because the light level on each plane is an important part of every set-up, a 
sensitive quick-reading light-measuring system had to be devised. The number of 
machine adjustments involved was so large that a master control sheet was laid out, 
giving complete operation information for each frame of film. As a final check 
before exposure, a periscope type of finder was devised so that the chief operator 
could give the set-up a visual check before each exposure. To write out the mas- 
ter control sheets it was necessary to develop a scene-planning group of artists and 
technicians to control and plan the use of the machine in creating the desired 

The results in giving increased power to animated motion pictures have been 
very satisfactory, as can be best illustrated by viewing the screen results. The 
adaptability of the multiplane technic to animation photography has proved to be 
so flexible that its complete possibilities will come only with experience. 

"A Method for Determining the Scanning Loss in Sound Optical Systems;" 
E. D. Cook, General Electric Co., Schenectady, N. Y., and V. C. Hall, Eastman 
Kodak Co., Rochester, N. Y. 

The usual methods of evaluating the frequency characteristic of sound records 
have been satisfactory for the determination of the required correction for overall 
losses. However, the losses due to aperture and optical effects are not known 
with sufficient precision to permit an inferior limit to be assigned to film loss only. 

The method described was chosen in connection with a high-fidelity develop- 
ment, and consists in comparing direct measurements made on images formed by 
contact printing of a geometrically shaped test-object on the film with measure- 
ments of frequency records made using the recorder optical system. While the 
results obtained can not be applied generally, the method is capable of segregating 
film loss from other losses for the specific conditions under which the test is con- 

"An Optical System for the Reproduction of Sound from 35-Mm. Film"; J. H. 
McLeod, Kodak Research Laboratories, Rochester, N. Y., and F. E. Altman, 
Hawk-Eye Works, Eastman Kodak Co., Rochester, N. Y. 

An optical system has been designed and tested for use in 35-mm. sound repro- 
ducers. It is the slitless type of optic, and gives a scanning image that is 0.001 
inch wide when used with an exciter lamp having a coil diameter of 0.055 inch. 
A toric lens is used to form a curved-line image of the filament of the lamp. This 
curved image is then re-imaged by a highly corrected objective lens of numerical 
aperture 0.28. The objective lens has inherent curvature of field, but this curva- 
ture is compensated for by the curvature of the line-image formed by the toric lens 
so that the final imag is flat. The toric lens also acts as a condenser lens to throw 

April, 1938] SPRING CONVENTION 483 

an image of the filament into the objective lens. Careful tests of samples show 
that the final image is flat, straight, and of uniform width and intensity. 

"Sound Recording by Color Modulation (Van Leer System)"; A. L. W. Wil- 
liams, Brush Development Co., Cleveland, Ohio. 

A method of recording sound is described in which advantage is taken of the 
variation in sensitivity of photographic film to light of different wavelengths. 
On standard film there is a portion of the sensitivity-wavelength curve in which the 
sensitivity changes linearly over a wide range with a small change in wavelength 
or color. An optical system and apparatus are described, designed to vary the color 
of the recording light over this narrow band and capable of wide-range recording. 
By this system very small deflection of the recording galvanometer is required, 
enabling a simple crystal oscilloscope to be used for the purpose. Chromatic 
aberration is almost eliminated. Large errors in exposure or development may 
easily be corrected so that minimum distortion occurs. 

Report of the Standards Committee; E. K. Carver, Chairman. 

The tentative drawings that have received initial and final approval by the 
Standards Committee have been published in the March issue of the JOURNAL of 
the Society. The uncompleted items at present under consideration are: 

(1) Drawings for standard cores for cine film. 

(2) Further consideration of the proper separation distance between the two 

halves of the push-pull sound-track. 
(5) Drawings of sprockets for 16-mm. sound-film. 

(4) Revision of the standard release print to correspond with the revisions 

made by the Academy. 

(5) Review and possible revision of the glossary of technical terms. 

(6) Carrying out actual tests on the new sprocket perforation for 35-mm. film, 

which, it is hoped, will displace the old Bell & Howell perforation. 

"An Ultraviolet Push-Pull Recording Optical System for Newsreel Cameras"; 
G. L. Dimmick and L. T. Sachtleben, RCA Manufacturing Co., Inc., Camden, 

Recent advances in sound recording technic, notably exposure with ultraviolet 
and the class B push-pull track form, are incorporated in a variable-width record- 
ing optical system of extraordinary compactness, for newsreel work. The overall 
dimensions are approximately 6 inches long by 4 inches wide by 3 7 /s high, and 
the weight complete is about 3 l /4 pounds. This compact form is made possible 
by the development of a new galvanometer window lens of very special form, and of 
an objective lens of 7.6 mm. E.F. and//2 speed which will cover the sound-track 
width satisfactorily. The power drain of the exposure lamp is 21 watts at 4.9 
volts, and the galvanometer input at full modulation is about 30 milliwatts. 

The class B push-pull track inherently provides ground-noise reduction with- 
out the use of special equipment. Response of the print at 6000 cps. is 3 db. below 
that at 1000 cps. with ultraviolet light, and 6 db. below it with white light. The 
turn of a lever and the reduction of the lamp current to 3 amperes prepare the sys- 
tem for white-light recording when battery power must be conserved and quality 
is less important. 


"Processing Ultraviolet Recording on Panchromatic Films"; J. O. Baker, RCA 
Manufacturing Co., Inc., Camden, N. J. 

The necessity in newsreel work of making the original sound recording on pan- 
chromatic film has always meant a serious sacrifice in quality and ground-noise 
ratio, as compared with the results that can be attained when sound is recorded on 
a separate film. While ultraviolet recording materially increases the fidelity of 
response, with panchromatic as well as with standard sound negative film, the low 
contrast and inherently high base fog of panchromatic film when processed for 
negative picture development produce noise and considerable reduction in vol- 
ume range. 

The track density on the panchromatic film is rather low, of the order of 1.0 to 
1.2, when recorded with a practical optical system for a single-film system. When 
this track is printed upon commercial release print stock the dense portion of the 
negative track will print through, producing a fog density in the clear portion of 
the printed track. This fog in the clear portion tends to produce noise and reduces 
the volume range. When the panchromatic negative and print are processed in 
accordance with commercial practice, the reduction in volume range is of the order 
of 6 decibels. 

Printing the panchromatic negative upon a high-contrast emulsion improves 
both the noise and volume range. Since the release prints must be on standard 
picture positive stock and not on high-contrast film, it is here proposed to make a 
master positive on high-contrast emulsion and to re-record from this to a standard 
sound negative, which would be used in the ordinary way to make the release 
prints. An improvement in release print ground-noise of 8 to 12 decibels is ob- 
tained by this method, and the volume range is increased by 6 decibels. Briefly, 
the proposed method is a means for increasing the density contrast of the final re- 
lease print track when the original is recorded on panchromatic film. 

"Design and Operation of Theater Loud Speaker Systems"; J. F. Blackburn, 
Lansing Manufacturing Co., Los Angeles, Calif. 

Although really satisfactory loud speakers have been commercially available 
for only a comparatively short time, all the essential elements of a good loud 
speaker have been at hand for many years, so that the reasons for the late appear- 
ance of suitable units must be sought in the economic rather than the technical 

The loud speaker with its horn and other adjuncts is considered analogous 
to the antenna and plate circuits of a radio transmitter. It is pointed out that 
probably only in acoustics and in radio transmission do we have to be so 
wavelength-conscious, since only in these cases do the wavelengths of interest 
range from very small to very large, compared with apparatus dimensions. 
This wide range at once denies the use of the types of simplifying assumptions 
that are so convenient in other fields, and introduces several sets of mutually con- 
tradictory requirements for the apparatus. To date, apparently no one has 
succeeded in fulfilling all these requirements in a single piece of apparatus, so that 
it becomes necessary to use multi-channel systems with appropriate frequency- 
dividing networks. 

One solution to the requirements just outlined is discussed in detail from the 
engineering point of view. The comparatively meager published design data are 

April, 1938] SPRING CONVENTION" 485 

reviewed and commented upon in the light of the author's experience with the 
units described. Some information is given regarding possible modifications of 
performance by minor changes in the units. Experiences in the application of 
these units in the field are discussed and suggestions are given to users. 

"Push-Pull Recording with the Light- Valve"; J. G. Frayne and H. C. Silent, 
Electrical Research Products, Inc., Hollywood, Calif. 

Push-pull recording on film is accomplished by means of a double light-valve 
having four ribbons. Distortions introduced by the recording medium which are 
represented by second-order harmonics balance out in reproducing, as do also the 
frequencies introduced by the action of the noise-reduction system. As a result, 
push-pull recording not only eliminates certain defects of conventional recording, 
but permits the application of new technics that allow further extension of the 
volume range and improvement in the naturalness in the final product. 

"The Educational Value and Preparation of U. S. Army Training Films"; 

R. T. Schlosberg, Capt., U. S. Army Signal Corps, Washington, D. C. 

Problems encountered, considered incident to the preparation of army training 
films, and teaching principles and their application to instruction through the 
medium of sound-films are discussed. The method employed and the criteria for 
the selection of subjects are outlined, as also the general method by which such 
film subjects are produced. 

"New Uses for Instructive Motion Pictures"; H. Roger, Rolab Photo-Science 
Laboratories, Sandy Hook, Conn. 

Problems are described that were encountered during the production of several 
motion pictures with sound for the New York State Department of Health. 
These films represent a type that has found new uses in instructing physi- 
cians and nurses, as well as the general public, in the treatment of pneumonia pa- 
tients. They represent a part of a nation-wide campaign program against the 
spread of pneumonia. One or two films will be demonstrated. 

"Making an Industrial Film"; J. A. Norling, Loucks & Norling Studios, Nevr 
York, N. Y. 

Industrial films can be classified as sales films, which are made for the purpose 
of putting a sales message across to the prospective consumer; sales-training 
films, which are devised to train salesmen and are not planned for public use; 
educational films, which may contain some sales-promotional material; advertis- 
ing films, which are usually very short bits released in theaters. Of the many in- 
dustrial pictures made in the last few years, by far the most important are those 
classified as sales, sales promotional and sales-training. 

Problems that arise in the production of these films are discussed. The in- 
creasing demand for color has set up many new problems for the producer of in- 
dustrial motion pictures and slide-films. Growing appreciation of high produc- 
tion quality among industrial clients has also increased the difficulties and ex- 
pense of the producer. These matters are touched upon but the main portion of 
the paper is devoted to one typical film a detailed case history of its making, from 
the original scenario to the ultimate use of the film in reaching the market. (The 
presentation will close with a demonstration reel.) 

486 SPRING CONVENTION [j. s. M. P. E. 

"An Industrial Visual Instruction Laboratory"; J. G. T. Gilmour, General 
Electric Co., Schenectady, N. Y. 

The history, methods of operation, equipment, and types of work are described 
of the section of the General Electric Co. that prepares, produces, and distributes 
the pictures used by the Apparatus and Supply Division of the Company. 

"A Higher-Efficiency Condensing System for Motion Picture Projectors"; 
F. E. Carlson, General Electric Co., Cleveland, Ohio. 

In motion picture projection optical systems for tungsten-filament sources, the 
condenser design is such that the source is imaged well ahead of the picture aper- 
ture. This position is dictated by considerations of uniformity of screen bright- 
ness. It is not the optimal position from the standpoint of utilization of light, for 
it entails losses at the aperture. At the best position for efficiency, the degree of 
brightness uniformity is inacceptable because of the non-uniform brightness of the 
source. The paper describes a method for reducing such losses without sacrificing 
picture quality. 

"A Water-Cooled Quartz Mercury Qamp"; E. B. Noel and R. E. Farnham, 
General Electric Co., Cleveland, Ohio. 

The structure of the water-cooled quartz mercury lamp, its operation, quality 
of radiation, brightness, and source size limitations are first described, followed by 
a discussion of the power-supply equipment, both a-c. and d-c. Applications of 
the lamp are as follows: 

(1) Motion picture projection, both with single lamps and with several 

(2) Motion picture photography, both black-and-white and color. This part 
of the paper tells also of an application to very high-speed motion picture photog- 
raphy. For black-and-white photography the lamp is quite satisfactory. For 
color work the relatively limited red radiation may call for external methods, 
either in the use of fluorescent reflectors or a highly red-sensitive emulsion, to make 
up for this deficiency. 

(3) Film printing. Because of the relatively high output in the blue-violet 
and ultraviolet regions this lamp may prove a very satisfactory source, especially 
where advantage is taken of the ultraviolet radiation. 

The following additional applications, of secondary interest to the motion pic- 
ture industry, are also discussed: photo-enlarging, photoengraving, and search- 

"Theory vs. Practice"; F. H. Richardson, Quigley Publishing Co., New York, 
N. Y. 

Attention is directed to the discredit heaped upon the splendid work accom- 
plished by scientific men in designing apparatus employed in projection, and upon 
the praiseworthy accomplishment of construction engineers in carrying those 
designs forward into completed equipments. Apparatus can not be made to func- 
tion efficiently in theaters while men are in charge who lack practical and theoreti- 
cal knowledge. The public, for the most part, is unable to form intelligent opin- 
ion as to where the fault for poor functioning lies, and almost invariably will credit 
it to imperfect equipment. Suggestions are offered looking toward placing thea- 

April, 1938] SPRING CONVENTION 487 

ter projection equipment in the hands of thoroughly capable men, to the end that 
the equipment may be made to produce results of which it is capable and to last a 
maximum length of time in service without excessively high operating expense. 

"Good Tools Pay for Themselves"; J. R. Prater, Congress Theater, Pabouse, 

The average projectionist does not equip himself with an ample supply of good 
tools, and the average theater management refuses to stock the projection room 
with anything more than a scant supply of tools of poor quality. After listing 
the tools that are known to be useful in the projection room the paper points out 
that were such tools available to the projectionist they would return their original 
cost in a relatively short space of time by enabling proper testing and alignment 
of equipment in addition to facilitating minor repairs of the equipment. 

"A Technic for Testing Photographic Lenses"; W. C. Miller, Paramount Pro- 
ductions, Inc., Hollywood, Calif. 

Different makes of lenses have different properties and characteristics which 
may render a lens ideal for one purpose and totally undesirable for another. Lenses 
of a given make and series often vary in quality among themselves. To obtain 
the best type of lens for a specific purpose it is necessary to subject the various 
makes obtainable to tests that will reveal the characteristics in such a way that 
they can be reduced to numerical quantities for comparison. Once the type of 
lens for a specific purpose has been chosen, it is of great importance to be able to 
select the best of that type from a group submitted by the manufacturer. 

Equipment and technic used in tests that make such discrimination possible 
are described. A few general hints and precautions are given that will aid in de- 
termining the characteristics most desirable for various purposes. Special em- 
phasis is placed upon the tests for lenses intended for use with standard 35-mni. 
equipment. It is a simple matter to apply the methods and principles to other 
classes of lenses. 

"Some Unusual Adaptations of 16-Mm. Equipment for Special Purposes"; 
J. L. Boon, Development Department, Eastman Kodak Company, Rochester, 
N. Y. 

A casual observer, looking over the existing standard amateur photographic 
equipment, would probably be of the opinion that there is little need of altering 
a camera to do a special job. However, closer observation of the various prob- 
lems that photography serves reveals that the standards of practice have been 
chosen merely to suit the needs of a common majority of users, and the minority 
are almost forgotten. Further observations show that an alteration to a stand- 
ard camera to make it fit a specific purpose usually precludes its usefulness for 
many of the purposes for which it was originally designed, and also its utility for 
other special purposes. 

An attempt has been made in this paper to make known some of the unusual 
adaptations of 16-mm. motion picture equipment, each to fulfill a definite pur- 
pose, and to show that industry is becoming more conscious of the utility of such 
photographic equipment as a tool in solving some of its problems. 


"A New 16-Mm. Projector"; H. C. Wellman, Camera Works, Eastman Kodak 
Company, Rochester, N. Y. 

The new projector is housed completely in aluminum die-castings, and to pro- 
vide quietness of operation, the pull-down gears are individually adjusted in as- 
sembly by means of an eccentric sleeve. To facilitate threading, the location of 
the pull-down claw is designated by the threading knob, the position of which 
can be detected by touch as well as by sight. Throwing a single lever engages the 
rewind mechanism and at the same time releases the lower reel. 

A threadlight is built into the projector, so positioned as to be most effective 
for threading the gate and the sprockets. The single control switch, a new and 
unique feature, has four positions: the first is the off position; the second turns 
on the threadlight so that the machine may be easily threaded in a darkened 
room; the third turns on the motor (the threadlight remains on so that the 
operator can momentarily see that the loops are properly formed and that the 
projector is functioning properly) ; and the fourth turns on the projection lamp 
and turns off the threadlight. 

"The Shrinkage of Acetate-Base Motion Picture Films"; J. A. Maurer and 
W. Bach, The Berndt-Maurer Corp., New York, N. Y. 

A simple direct-reading film-shrinkage gauge has been constructed with which 
shrinkage readings may be made in a few seconds. The accuracy of the instru- 
n ent is such that the maximum variation in a series of readings made upon a par- 
ticular film will not be more than 0.02 per cent of the predetermined length mea- 
sured. Readings have been taken systematically with this instrument over a pe- 
riod of five months to determine the shrinkage behavior of acetate-base films under 
various conditions of storage and use. 

The results indicate that the safety-film base made by each of the three Ameri- 
can manufacturers has a characteristic value of shrinkage that is ordinarily 
reached within a few days after processing. Subsequent shrinkage is slow but con- 
tinuous over a long period of time. The ultimate value of shrinkage is of the order 
of 1.25 per cent except in the case of films that have been projected many times on 
projectors using high-wattage lamps. The bearing of this shrinkage information 
upon equipment design is discussed briefly. 

"A Criticism of the Proposed Standard or 16-Mm. Sound-Film"; J. A. Maurer 
and W. H. Offenhauser, Jr., The Berndt-Maurer Corp., New York, N. Y. 

It has been proposed that the standard dimensions of 16-mm. sound prints be 
changed, principally by widening the sound record and scanned areas. The ques- 
tion is reviewed from the standpoint of the cumulative effects of film shrinkages 
and mechanical inaccuracies in the steps leading to the final sound print and in the 
projection of that print, following the method described by R. P. May in the 
April, 1932, JOURNAL. 

A film having sound records of various widths will be demonstrated to support 
the contention that increased width of sound-track is not needed, and that if any 
change from the present standard is to be made, it should be in the direction of a 
narrower track to provide a wider margin outside the sound-track and a wider 
safety area between the sound-track and the picture. 

April, 1938] SPRING CONVENTION 489 

"A Continuous Optical-Reduction Sound-Printer"; M. G. Townsley, Bell & 
Howell Co., Chicago, 111. 

Optical-reduction printing from 35-mm. negative to 16-mm. positive has come 
into wide use. A new printer has been developed for making optical-reduction 
prints. The printer departs from conventional design in that the film rolls are 
horizontal, making possible oil-damped filters and flood lubrication without fric- 
tion-producing oil-seals. The printer operates in either direction and stops auto- 
matically at the end of the negative. A three-phase, 220-volt synchronous motor 
drives the main worm shaft from which all the working parts are driven at a print- 
ing speed of 60 feet of 35-mm. film per minute. Uniform film motion is achieved 
by a heavy flywheel and independently filtered drive to each film drum . 

The self-contained optical unit produces on the 16-mm. film an image of the 
35-mm. track moving in synchronism with the 16-mm. film, with longitudinal and 
transverse mangifications of 0.40 and 0.84, respectively. Provision is made for 
printing masking lines at the edge of the track. A 10-volt TVs-ampere d-c. lamp is 
operated from a pair of 6-volt storage-batteries and a full-wave charger. Lamp 
current is controlled by a rheostat and ammeter. 

"An Automatic Camera Timer for Time-Lapse Cinematography"; H. Roger, 
Rolab Photo-Science Laboratories, Sandy Hook, Conn. 

Ever since the invention of motion picture, time-lapse cinematography has been 
used extensively to speed up slow actions. More or less complicated devices have 
been constructed, mostly by the cameraman himself, to take single exposures at 
various time intervals. In modern cinematography, especially in the industrial 
and scientific field, time-lapse photography has found a great many new uses in 
recording slow processes. The camera timer described in this paper operates 
not only the camera but also the light, in synchronism with the camera shutter. 
The timer is the result of more than twenty years of practical experience in this 

"A New Framing Device for 35-Mm. Projectors"; H. A. DeVry, Herman A. 
DeVry, Inc., Chicago, 111. 

This device embodies a unique application of the silent chain drive to the mo- 
tion picture mechanism, in such a way that the up or down movement of the 
film effected by the framer is accomplished without disturbing the synchronism 
between the shutter and the intermittent. Also, since the framing is done by an 
overhanging arm built directly onto the intermittent, the intermittent moves 
only rotationally, and remains always so close to the aperture that there is no room 
for buckling of the film. In fact, it is impossible for any buckling to occur due 
to framing. 

"A Film Cement Pen"; R. J. Fisher, Rochester, N. Y. 

The purpose of this device is to make the application of film cement in splicing 
film easier and neater, and allow no waste of cement by spilling or evaporation. 
It replaces the bottle, brushes, medicine droppers, etc., and is a time-saving ele- 
ment where it is necessary to make many splices, as in film exchanges, studios, and 


"New Piezoelectric Devices of Interest to the Motion Picture Industry"; 
A. L. Williams, The Brush Development Co., Cleveland, Ohio. 

Devices discussed are : (a) phonograph pick-up with uniform response (without 
further compensation) 30 to over 10,000 cps. with forces so low that it will repro- 
duce this range from soft direct recordings with negligible wear; (&) record cutter 
which, used in conjunction with the pick-up, will record over same range; (c) 
high-fidelity headphones reproducing to over 10,000 cps. with high sensitivity 
and high impedance (over 7500 ohms per pair); (d) unidirectional microphone 
(changeable at will to bidirectional or non-directional) using ribbon pressure 
gradient unit and sound cell pressure unit. 

"Characteristics of Supreme Panchromatic Negative"; A. W. Cook, Agfa 
Ansco Corporation, Binghamton, N. Y. 

The new panchromatic negative film is compared with earlier types of super- 
sensitive material. It has a light-sensitivity twice as great as that of Superpan. 
This permits a 50-per cent reduction in set lighting, or the use of a smaller lens 
aperture for gaining greater depth of field with undiminished illumination. Rela- 
tive color-sensitivity is substantially unaltered. The film is doubly coated, with 
two emulsion layers superposed upon a gray antihalation layer between the emul- 
sion and the support. Despite increased sensitivity, Supreme negative has 
equally good keeping qualities, finer grain, and lower developing fog than Super- 
pan. Development characteristics are similar and no alteration of laboratory pro- 
cedure normally employed for typical supersensitive materials is suggested, al- 
though the long scale of the film allows great latitude in development. Extremes 
of light and shade beyond limits imposed by earlier supersensitive materials can 
be recorded faithfully, as indicated by the long straight-line portion of the charac- 
teristic curve, a very short toe, and a shoulder falling in the region of densities far 
beyond those encountered in practice. These advantages are reflected in the 
quality of negatives taken under adverse lighting conditions. 

"A New Indicator for Sound-Level Measurements"; S. K. Wolf and S. J. 
Begun, Acoustic Consultants, Inc., New York, N. Y. 

This device consists of a long glass tube approximately 18 inches long and one 
inch in diameter. In the tube are three electrodes, one of which extends the en- 
tire length of the tube, and a mixture of inert gases at a very low pressure. The in- 
side of the glass is coated with a material that fluoresces under the ultraviolet and 
positive ion bombardment, due to a constant voltage applied to the electrodes 
of the tube. The coating is of three different types, each of which will fluoresce 
a different color green, blue, and red. The "blue" extends for seven inches at 
the lower end of the tube, next the "green" for four inches, and finally the "red" 
for the upper remaining seven inches. 

A voltage of 250 volts is placed across the two small electrodes to produce a 
striking voltage and establish a zero point on the tube. Then the alternating 
voltage is applied to the long electrode from the output of a voltage amplifier 
to which a microphone is attached to pick up the sound under observation. 
The range of the tube is approximately 70 db. When the signal is too weak, the 
"blue" portion of the tube lights up; if the signal is brought up to a higher level 
the "green" portion lights up, representing the 3-db. change required to modulate 

April, 1938] SPRING CONVENTION 491 

a broadcast station from 65 to 85 per cent. If the signal is still stronger the "red" 
portion of the tube lights. Approximately 250 volts a-c. are required to operate 
the tube over its full range. The tube can be calibrated to read directly on a 
decibel scale, and by using a special type of logarithmic amplifier, the scale is 
linear. The great advantage of the tube is that the entire audio intensity range 
is on one scale and no switching scales need be done in operating this instrument. 
As a practical example of the use of the tube, the blue region may indicate that 
a speaker's voice is too low, the green that it is satisfactorily strong, the red that 
the voice is too strong. 

"The Resonoscope"; S. K. Wolf and L. B. Holmes, Acoustic Consultants, Inc., 
New York, N. Y. 

The "resonoscope" employs a special cathode-ray oscillograph in conjunction 
with a standard set of musical frequencies the twelve notes of the chromatic musi- 
cal scale, produced by twelve electrically driven tuning forks, which synchronize 
an oscillator in step with them. This oscillator provides a horizontal sweep cir- 
cuit for the cathode-ray tube. A voltage amplifier picks up music or any single 
musical tone, through a crystal microphone, and energizes the vertical plates of 
the cathode-ray tube. This gives a visual picture of the wave-form of the musi- 
cal note under observation. If the note is of the same pitch (or frequency) as the 
standard, or any harmonic of it, the wave-form will appear stationary on the screen 
of the cathode-ray tube. If the note is flat, or lower in pitch, than the horizon- 
tal sweep standard, the wave-form will appear to be moving to the left; if 
higher in pitch than the standard, or sharp, the wave-form will move toward the 
right. The movement indicates to the musician whether he is playing in tune 
or is sharp or flat. The speed with which the wave-form moves across the screen 
is an indication of the extent to which the instrument is out of tune. 

Any of the twelve standard frequencies in the instrument may be selected by 
turning a control on the panel of the instrument, and each setting of the control 
accommodates all octaves of the particular note. One of the special features of 
the circuit is that the horizontal sweep circuit is automatically changed in fre- 
quency to compensate for the change in frequency in going from one note to an- 
other. This allows the sweep circuit to be synchronized at all times by the stand- 
ard frequency of the tuning forks and assures the observer that the number of 
wave-forms on the screen of the cathode-ray tube is a direct indication of the oc- 
tave he is playing or tuning. The frequencies of the standard chromatic scale 
are calculated for a true tempered scale, which has the most practical use for all 
types of tuning. The pitch of the scale is 440 cycles per second for A above 
middle C, this being the international pitch for tuning. This pitch is the one 
being used in the present models but any pitch can be had by substituting a new 
set of standards. 

"A Roller Developing Rack with Stationary Drive"; C. E. Ives, Kodak Re- 
search Laboratories, Rochester, N. Y. 

In a previous paper a rack was described that provided for continuous motion 
of a 200-ft. length of motion picture film during processing but could be used with 
the rack-and-tank equipment. The purpose of this roller rack was to give a type 
of treatment in processing essentially similar to that given by a continuous ma- 


chine while retaining the features of batch equipment that are helpful in experi- 
mental processing. 

The rack previously described included a built-in driving motor and reduction 
gear, an arrangement that was most feasible for a single unit. With more exten- 
sive use it became desirable to have multiple units operated from stationary 
drives at the tanks and at the loading and unloading stations. 

A new design has been worked out in which the weight of the rack was reduced 
greatly by the use of stationary drives. Further reduction in weight was attained 
by the substitution of tensioning springs for the weighted supporting beam asso- 
ciated with the movable lower shaft in the earlier model. This shaft was 
mounted upon the frame by lever arms in such a way as to use the torsional rigidity 
of the shaft itself to maintain it parallel to the upper shaft while allowing it the 
necessary vertical movement. 

"A New Projector Mechanism"; H. Griffin, International Projector Corp., New 
York, N. Y. 

This new projector is provided with synchronized front and rear shutters op- 
erating in the same direction and providing considerably greater screen illu- 
mination; an automatic fire-shutter safety trip for fire prevention; a Bijur one- 
shot oiling system to provide positive lubrication under pressure, together with 
ball bearings having sealed lubrication for extremely long service; heavier film- 
gate construction, the entire unit being readily removable for cleaning and having 
adjustable tension devices and locking positively both in the open and closed posi- 
tion; readily temovable film -trap having edge-guiding means and provision for 
easily cleaning and replacing worn film runners ; a new ring-type fire-shutter gov- 
ernor; easier threading facilities; new automatically positioned threading lamp; 
illuminated pearl gray enameled interior; and other distinctive improvements. 

"New Safety Switch for Motion Picture Projection Rooms"; E. R. Morin, 
Department of State Police, Hartford, Conn. 

An emergency switch has been designed for projection rooms, which in the 
event of a fire simultaneously starts or speeds up the ventilating fans, and turns 
on the auditorium lights. Details are given of the safety requirements for the 
construction of theater projection rooms in the State of Connecticut. 

"A Solution of the Galvanometer Window Lens Problem in Recording Optical 
System Design"; G. L. Dimmick and L. T. Sachtleben, RCA Manufacturing Co., 
Inc., Cam den, N. J. 

In the design of the variable-width recording optical system, the lens that im- 
ages the aperture upon the slit should, ideally, be located at the galvanometer mir- 
ror. Early systems employing a vertical cutting edge permitted, with fair success, 
the use of a simple lens close to the mirror, but the stigmatic image thus obtained 
ruled the method out when oblique cutting edges were adopted. Resort was then 
made to a simple achromatic lens located axially between the aperture and gal- 
vanometer mirror. 

The design of a very compact system for newsreel work has required that the 
lens be again located at the galvanometer mirror. This makes it necessary for the 
light to pass obliquely through the lens both before and after reflection from the 

April, 1938] SPRING CONVENTION 493 

mirror. Both a simple lens and a corrected lens have been designed to meet this 
requirement, and it is found that definite advantages in the way of image quality 
inhere in such a lens when the design is properly executed. 

"A Study of Processing Conditions for the High-Resolution Sound Recording 
Emulsions"; J. O. Baker, RCA Manufacturing Co., Inc., Camden, N. J. 

The high-resolution recording emulsion described in the January, 1938, issue of 
the JOURNAL has been found satisfactory for recording as a negative provided that 
a sufficiently high density is obtained. With ultraviolet recording on the stand- 
ard emulsions, the best value of print density is of the order of 1.4 for a negative 
density of 1.9. 

The inherently low image-spread of the high-resolution emulsion requires a 
higher negative density for the same print density of 1.4. The negative density is 
of the order of 2.2 to 2.5. The noise is thereby reduced for two reasons: first, 
the inherent noise of the high-resolution emulsion is lower than that of the stand- 
ard emulsions; and second, the higher negative densities give less trouble from the 
so-called "pin-hole" effect. 

General practice at the present time is to make a master sound positive of the 
selected takes from which a re-recorded negative is made for use in the production 
of release prints. The paper discusses the processing conditions for the high-reso- 
lution emulsion when used as (a) an original negative, (6) as a master positive, and 
(c) as a re-recorded negative for final printing onto the standard release print posi- 

"A New Sound System"; G. Friedl, Jr., International Projector Corp., New 
York, N. Y. 

A brief review is given of the advanced features of the new Simplex sound sys- 
tem, and the considerations involved in developing a high-quality system for 
small as well as for large theaters are outlined. The engineering requirements of 
high-quality reproduction are set forth, and the methods employed for meeting 
these requirements even in the smallest system are explained. 

The development is outstanding because of its very low noise level, which in- 
sures an effectively wide volume range. The advantages of wide frequency range 
are preserved by the special care given to the reduction of flutter. The power 
of even the smallest system is sufficient to reproduce faithfully the latest improved 
types of recordings, such as the "Hi-Range" prints. The system employs a re- 
fined type of rotary stabilizer mechanism, with provision for dual track reproduc- 
tion, such as from push-pull or stereophonic recordings. Special facilities are pro- 
vided for mounting and adjusting the projector mechanism. Change-over 
controls are of the electronic type employing grid -biasing circuits so as to eliminate 
switch contacts and mechanically interlocked controls. Standardized power 
amplifiers of 15-watt capacity with extremely low limits of harmonic distortion 
are used singly or in parallel for various system combinations. Two-way loud 
speaker systems are employed, with special switching facilities that simplify 
checking loud speaker units as well as amplifier characteristics. 

"The Properties and Applications of Ozaphane"; J. Holloway, The Holloway 
Co., New York, N. Y. 
Chemical and mechanical differences between Ozaphane and gelatin emulsion 

494 SPRING CONVENTION [j. s. M. P. E. 

films are discussed. A report is made of accelerated life tests conducted by the 
New York Testing Laboratories and the U. S. Bureau of Standards. The dupli- 
cating properties of Ozaphane are discussed, and reference is made to the following 
applications: sound-track, home phonographs, radio broadcast, organ recordings, 
etc.; microphotography trends, resolution of Ozaphane; color transparencies; toy 
film, in black-and-white and in color. 

The diazo dye process is treated as applied to bases other than cellophane ; sur- 
face and complete sensitization. A spectrographic analysis is given of diazo dye- 
stuffs and it is shown how projection and sound-track utilizations are affected. 
(Samples of film will be projected and a demonstration will be made of a home 
phonograph using Ozaphane film.) 

"Tracing-Distortion in Sound Reproduction from Phonograph Records"; J. A. 
Pierce and F. V. Hunt, Cruft Laboratory, Harvard University, Cambridge, Mass. 

When the spherical tip of an ideal reproducer stylus slides over a warped groove 
surface having a sinusoidal profile, the traced curve is not exactly sinusoidal. An 
analysis of the harmonic content of the traced curve, similar to that given by Di 
Toro (J. Soc. Mot. Pict. Eng., Nov., 1937) but avoiding his approximations, is 
directly applicable to reproduction from vertical-cut records. These results may 
be applied to reproduction from lateral-cut records by taking the original groove 
surface as inclined approximately 45 degrees from the horizontal, projecting the 
traced curve upon the horizontal and vertical planes, and adding in proper phase 
the guidance of the stylus tip by both sidewalls. It is shown that there is a 
residual vertical component of stylus motion ("pinch" effect) and complete can- 
cellation of all even harmonics in the tracing distortion. Computation of the re- 
maining odd harmonics indicates that, when the ideal lateral-cut reproducer char- 
acteristics include ideal "following" for vertical motion at signal frequency, a 
lateral-cut record may be reproduced with one-fourth to one-tenth the rms. dis- 
tortion of a similarly recorded vertical-cut record. These results are displayed 
for convenient reference by contours of constant distortion upon a universal chart, 
the dimensionless coordinates of which characterize any recording condition and 
allow immediate specification of the maximum permissible recorded amplitude, 
maximum predistortion of the frequency characteristic, and the required clear- 
ance angle of the recording stylus. 

"Multiple-Channel Recording"; H. G. Tasker, Universal Pictures Co., Inc., 
University City, Calif. 

Multiple-channel recording is a device for achieving needed flexibility at the 
time of dubbing or re-recording orchestral music presented as such in the picture. 
If producers, directors, and film editors could predict for the music and sound de- 
partments which portions of the orchestra would be seen from which angles in the 
finally edited picture, or if the editing could be completed before the music was 
recorded, there would be less merit in multiple-channel recording. 

The reverse is true: The music is recorded first, the musicians photographed 
later, synchronizing their movements to a play-back of the record. Meanwhile, 
the pictorial treatment has taken partial shape in the minds of producer and di- 
rector. Still later it takes final shape in the hands of the film editor. Sound and 
action are then placed in the hands of the sound department for dubbing, but 

April, 1938] SPRING CONVENTION 495 

it is then far too late to do more than an ineffectual raising and lowering of volume. 
The violins or the woodwinds can not be lifted above the surrounding sections to 
match a close-up of the picture. 

Multiple recording, meaning the provision of a separate recording channel, 
complete with microphone, amplifier, recording machine, etc., for each im- 
portant section of the orchestra (usually six) and all propelled in synchronism, 
provides an excellent solution of the problem. The resulting multiplicity of 
sound-tracks (recorded, of course, in advance of the photography) will afterward 
provide the dubbing mixer with the means of easily blending a final sound-track 
that will be wholly in keeping with the final edition of the picture. The applica- 
tion of this method to the recent production 100 Men and a Girl is described. The 
use of "close-mix" tracks, separate vocal tracks, etc., in conjunction with multiple 
recording is also described. 

"Application of Non-Linear Volume Characteristics to Sound Recording"; 
J. O. Aalberg and V. G. Stewart, RKO Radio Studios, Inc., Los Angeles, Calif. 

The advisability of using a non-linear volume characteristic in dialog record- 
ing is discussed. In this connection consideration is given to the following points: 
(0) the difference of level existing between the original and reproduced speech; 
(b) the advantages of a system in which manual monitoring can be confined to 
overall level correction rather than to momentary peaks; (c) the advantage of 
limiting the range of all except trained voices to assure the highest possible intel- 
ligibility. An analysis is then made of the various types of compression possible 
and a terminology is developed. 

Consideration is given to the type of device most applicable to motion picture 
recording. The electrical circuits and operating characteristics of a compressor 
that has been in commerical service for 18 months are discussed. Practical re- 
sults and advantages obtained by the use of the device during this period are 
analyzed and the possibility of additional applications is indicated. 

"The Philips-Miller Method of Sound Recording"; R. Vermeulen, N. V. 
Philips Gloeilampenfabrieken, Eindhoven, Holland. 

The first attempt at mechanographic recording and reproducing was made as 
early as 1891 but no successful solution of the problem was applied until the inven- 
tion of J. A. Miller in 1931. The principle of this invention is described as inertia- 
free magnification. After the introduction of the principle, the inventor cooper- 
ated with the research laboratories of the N. V. Philips Company, of Eindhoven, 
Holland, in order to solve the problems involved in bringing the system to commer- 
cial use. 

The method of obtaining a mechanical amplification of forty times is described 
and illustrated. The mathematical and theoretical advantages of the sytem over 
photographic methods are discussed, as also the film or tape that has been spe- 
cially prepared for this process of recording. Some of the difficulties or precau- 
tions that are peculiar to the system and are new to the art of recording, relating 
to the cutting instrument, cutting material, and the coating of the film, are de- 
scribed. One type of recording machine is described with drawings of the most 
interesting mechanical parts. A bibliography of articles on the subject is ap- 


"Electrical Networks for Sound Recording"; F. L. Hopper, Electrical Research 
Products, Inc., Hollywood, Calif. 

Electrical networks are employed in sound recording for modifying and limiting 
the frequency-response characteristic. The necessity for their use, application, 
and design are described. Particular emphasis is placed upon the constant-resis- 
tance type of structure. 

"The Application of Electrical Networks to Sound Recording and Reproducing"; 
H. R. Kimball, Metro- Goldwyn- Mayer Studios, Culver City, Calif. 

The use of electrical networks with recording and reproducing systems to ac- 
complish beneficial results has been steadily increasing. Two types of networks 
are in general use, namely, wave-filters and attenuation equalizers. This paper 
discusses in some detail the use of these networks with sound systems as reflected 
by present practices and later presents practical data for engineering the networks 
with a minimum of time and effort. The uses to which attenuation equalizers are 
put divide these networks into two general classes: first, fixed equalizers to pro- 
vide fixed equalization for sound channels; and, second, variable equalizers to 
provide means for varying at will the relative amplitudes of the frequency 
components of sound signals. Although the means for engineering variable net- 
works is far from being ideal, the review given in the paper of present practices 
should be valuable. 

"Silent Gasoline Engine Propelled Apparatus"; J. E. Robbins, Paramount 
Pictures, Inc., Hollywood, Calif. 

Problems are discussed connected with the design, construction, and operation 
of electrical generators and water pumps running under full load sufficiently 
silently to permit satisfactory sound recording. The units described were the result 
of demands for silent power equipment for making shots on boats, trains, bus in- 
teriors, inaccessible canyons, etc. As an example of what is sometimes required, 
one of the largest units was installed in the hold of a windjammer used throughout 
the Paramount Production, Souls at Sea, and although the microphone was at 
times directly above (approximately 30 feet) the spot occupied by the generator, 
no noises were picked up by the sound recording equipment. 

Four units are described, namely, one 144-kw. Hispano Suiza, one 57-kw. 
Lincoln Zephyr, and one 41-kw. Ford V-8 generator, and one high-pressure Ford 
V-8 water pump. In each case the entire mechanical unit is rubber-mounted on 
a sub-frame within a semi-airtight compartment constructed of an outer shell of 
22-gauge auto-body steel, four inches of sound-absorbing material with an inner 
lining of asbestos cloth. The entire exhaust system is water-cooled, employing 
special mufflers also housed within the case. One radiator, mounted outside, 
cools the water for the engine as well as the exhaust. All are practically automatic 
in operation, with electrical governors, temperature regulators, etc. The ma- 
chines have been in operation approximately fifteen months and have required 
very little service other than normal maintenance. 

"Variable-Matte Control (Squeeze-Track) for Variable-Density Recording"; 
G. R. Crane, Electrical Research Products, Inc., Hollywood, Calif. 

A review of the relation between the width of variable-density sound-track and 
the signal-to- noise ratio indicates the advantages to be gained by applying a vari- 

April, 1938] SPRING CONVENTION 497 

able matte to the sound-track. To provide this facility, a sound-track matting 
system has been developed for application to existing standard studio equipment. 
By means of selsyn-type motors, a foot-operated control unit drives an indicating 
meter, an attenuator, and a masking device on the recorder. A new condenser 
lens assembly is used on the recorder and the system may be used for either single 
or push-pull recording. 

"Permanent-Magnet Four-Ribbon Light- Valve for Portable Push-Pull Re- 
cording"; E. C. Manderfeld, Electrical Research Products, Inc., Hollywood, 

The general adoption of the push-pull recording technic has necessitated pro- 
viding adequate modulating equipment for portable recording channels. The 
four-ribbon permanent-magnet light-valve herein described is an important unit 
in this equipment. It has been designed to provide the smallest practicable me- 
chanical structure without sacrificing the operating and maintenance advantages 
possessed by the larger type of valve used in fixed channels. The magnetic field 
of the valve is provided by permanent magnets. The individual ribbons are so 
mounted as to allow spacing and tension adjustments at any time. 

"Overload Limiters for the Protection of Modulating Devices"; R. R. Scoville, 
Electrical Research Products, Inc., Hollywood, Calif. 

Two types of volume-limiting devices are discussed. The first type automati- 
cally limits the envelope of a signal to a predetermined maximum amplitude in 
such manner that harmonics are not generated. A time factor is incurred wherein 
the envelope amplitude is changed when the limiting value is approached. A 
second type of volume limiter acts instantaneously to prevent excessive signal 
amplitudes, and is used primarily for the protection of equipment against damag- 
ing signals and where such odd harmonics as are generated during the limiting 
period can be tolerated. Equipment of this type is described in detail and com- 
pared with limiters of the first type. 



Details of the forthcoming Convention to be held at Washington, D. C., with 
headquarters at the Wardman Park Hotel, are given in the preceding section of 
this issue of the JOURNAL. The dates will be April 25th to 28th, inclusive, and 
many interesting features are being arranged by the Convention and Papers 
Committees under W. C. Kunzmann, Convention Vice- President, and J. I. 
Crabtree, Editorial Vice- President. 


Two meetings of the Committee were held on February 17th and March 24th, 
and two additional meetings are planned before the Spring Convention, in order 
to complete the semi-annual report. The report this year will be very compre- 
hensive, presenting an analysis of the data obtained through the theater survey 
conducted during the past year or so and announced in previous issues of the 
JOURNAL. Charts are being prepared showing the ranges of auditorium shapes, 
projection angles, and other important features of theater design encountered 
in the field. In addition, considerable work is being done in drawing up a set 
of recommendations to the National Fire Protection Association for revising the 
N. F. P. A. "Regulations for Nitrocellulose Motion Picture Film." Other Sub- 
Committees also will be prepared to report in addition to those mentioned. 


At a meeting held on March 15th in the meeting rooms of the Western Society 
of Engineers, Chicago, Mr. Charles Herbst, Jr., of RCA Manufacturing Com- 
pany, presented a paper on "Loud Speaker- Developments." The presentation 
included a discussion of amplifier distortion and compensation, and sound-head 
analysis. The next meeting is scheduled for April 12th. 


At a meeting held on March 23rd in the studio of RCA Photophone, Inc., 
New York, N. Y., Dr. S. J. Begun presented a "Symposium on the Recording 
and Reproduction of Speech." Particular attention was given to electromagnetic 
recording and reproducing, and a demonstration was given of direct recording 
and playback. The "resonoscope," an instrument for comparing true voice 
pitch values, was demonstrated, as well as an automatic volume indicator for 
speech level. 




At a recent meeting of the Admissions Committee, at the General Office of 
the Society, the following applicants for membership were admitted to the 
Associate grade: 


2142 J /2 Beachwood Terrace, 

Hollywood, Calif. 

969 So. Serrano Ave., 
Los Angeles, Calif. 
1010 Third St., 

Alamosa, Colo. 
2996 Lawrence St., 

Detroit, Mich. 
2869 Grand Concourse, 

New York, N. Y. 

1110 E. Palmer Ave., 

Glendale, Calif. 
909 Vine St., 

Louisville, Ky. 

81 Devonshire Rd., 
Forest Hill, S. E. 23, 
London, England. 

18 Rue de Sucy 

Seine et Oise, 
Paris, France. 

KUHN, J. G. 

1345 Winchester Ave., 

Glendale, Calif. 
410 Ogden St., 
Denver, Colo. 


McArthur Theatre Equipment Co., 
2501 Cass Ave., 
Detroit, Mich. 


4738 Tobias St., 

Van Nuys, Calif. 
725 Longwood Ave., 
Los Angeles, Calif. 
Moss, G. 

1311 Second St., 
Alamosa, Colo. 

208 W. Montcalm St., 

Detroit, Mich. 

5 Arnold Court, Truro Rd., 
Bowes Park, 
London N-22, England. 
2055 Harrison Ave. 
New York, N. Y. 

In addition, the following applicants have been admitted by vote of the Board 
of Governors to the Active grade: 


Electrical Research Products, Inc. 
250 West 57th St., 
New York, N. Y. 


Bell & Howell Co., 
4045 N. Rockwell St., 
Chicago, 111. 


2237 Mandeville Canyon Rd., 
Culver City, Calif. 

S. M. P. E. 


These films have been prepared under the supervision of the Projection 
Practice Committee of the Society of Motion Picture Engineers, and are 
designed to be used as precision instruments in theaters, review rooms, 
exchanges, laboratories, factories, and the like for testing the perform- 
ance of projectors. 

Only complete reels, as described below, are available (no short sections 
or single frequencies). The prices given include shipping charges to all 
points within the United States ; shipping charges to other countries are 

35-Mm. Sound-Film 

Approximately 500 feet long, consisting of recordings of several speak- 
ing voices, piano, and orchestra; buzz-track; fixed frequencies for focus- 
ing sound optical system; fixed frequencies at constant level, for de- 
termining reproducer characteristics, frequency range, flutter, sound- 
track adjustment, 60- or 96-cycle modulation, etc. 

The recorded frequency range of the voice and music extends to 10,000 
cps. ; the constant-amplitude frequencies are in 15 steps from 50 cps. to 
10,000 cps. 

Price $37.50 each, including instructions. 

35-Mm. Visual Film 

Approximately 500 feet long, consisting of special targets with the aid 
of which travel-ghost, marginal and radial lens aberrations, definition, 
picture jump, and film weave may be detected and corrected. 

Price $37.50 each, including instructions. 

16-Mm. Sound-Film 

Approximately 400 feet long; contents identical to those of the 35-mm. 
sound-film, with the exception that the recorded frequency range ex- 
tends to 6000 cps., and the constant-amplitude frequencies are in 11 
steps from 50 cps. to 6000 cps. 

Price $25.00 each, including instructions. 

16-Mm. Visual Film 

An optical reduction of the 35-mm. visual test-film, identical as to 
contents and approximately 400 feet long. 
Price $25.00 each, including instructions. 







Volume XXX MAY, 1938 Numbers 


A Method of Enlarging the Visual Field of the Motion Picture 


A Horn Consisting of Manifold Exponential Sections 

H.F.OLSON 511 

Scoring-Stage Design M. RETTINGER 519 

Recent Developments in Background Projection 


Sensitivity Tests with an Ultra-Speed Negative Film 

P. H. ARNOLD 541 
Reduction Potential and the Composition of an MQ Developer 

Infrared Absorption by Water as a Function of Temperature of 

Radiator A. H. TAYLOR 568 

Golden Jubilee Anniversary of the Motion Picture Industry . . . 

New Motion Picture Apparatus 

New Ideas in Mobile Sound Recording Equipment 


A New Motion Picture Camera Crane 


A Sound- Film Phonograph. .D. CANADY AND V. A. WELMAN 591 
Precision All-Metal Reflectors for Use with Projection Arcs 

C. E. SHULTZ 594 

A Device for Cleaning Sound-Track during Projection 

R. J. FISHER 597 

Flash Fire- Valve R. J. FISHER 600 

A Portable Loose-Sheet Microphotographic Camera 

R. H. DRAEGER 601 

Current Literature 605 

Book Reviews 608 

Officers and Governors of the Society 610 

Committees of the Society 613 

Society Announcements 618 

Abstracts of Papers from the Washington Convention 620 





Board of Editors 
J. I. CRABTREE, Chairman 



Subscription to non-members, $8.00 per annum ; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West-Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1938, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 


President: S. K. WOLF, RKO Building, Rockefeller Center, New York, N. Y. 

'Past-President: H. G. TASKER, Universal City, Calif. 

Executive Vice-President: K. F. MORGAN, 6601 Romaine St., Los Angeles, 


** Engineering Vice-President: L. A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: E. A. WILLIFORD, 30 E. 42nd St., New York, N. Y. 
Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: J. FRANK, JR., 90 Gold St., New York, N. Y. 
Treasurer: L. W. DAVEE, 250 W. 57th St., New York, N. Y. 


*J. O. AALBERG, 157 S. Martel St., Los Angeles, Calif. 
*M. C. BATSEL, Front and Market Sts., Camden, N. J. 
**R. E. FARNHAM, Nela Park, Cleveland, Ohio. 
*G. FRIEDL, JR., 90 Gold St., New York N. Y. 
*A. N. GOLDSMITH, 444 Madison Ave., New York N. Y. 
**H. GRIFFIN, 90 Gold St., New York, N. Y. 

**A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 
*S. A. LUKES, 6145 Glenwood Ave., Chicago, 111. 
*Tenn expires December 31, 1938. 
**Term expires December 31, 1939. 



Summary. Recent trends toward the smaller sized motion picture audience indi- 
cate that new considerations can be given to the possibility of a larger and differently 
shaped screen, retaining the 35-mm. film. The screen is pictured as completely occu- 
pying the entire forefront of the motion picture auditorium, assuming a space stage 
instead of an artificially framed picture. 

The first step toward making pictures lifelike was to add the 
effect of motion. Then sound was added, and now natural color has 
become a factor necessary to enhance the effect of realism of the motion 
picture. There are still to be considered, however, two more elements 
required to render the picture completely lifelike. 

The first of these, that of obtaining a sense of depth or relief, is 
occupying the minds of many at the present time. The second is an 
important factor that has not received sufficient attention in the 
past; that is, making the picture appear to fill the field of view of the 
spectator in the theater, so that the spectator is no longer "picture 
conscious." Rather he should be made to feel that what is being un- 
folded before his eyes is very much the same as his natural field of 
vision in real life. 

There are two reasons why this effect can not be achieved under 
the present conditions of motion picture projection. One is the 
limitation of size of screen and motion picture film, and the other is 
the artificiality of the black border, which sharply cuts off the edges 
of the picture. The motion picture screen as now presented occupies 
only a small portion of the field of vision of the spectator sitting in 
the theater. Surrounding the screen is not only the black masking 
but also wall and ceiling surfaces unrelated to the picture, the illu- 
mination levels of which are disconcerting to the viewer and irrelevant 
to the illumination of the screen. 

* Presented at the Fall, 1937, Meeting at New York, N. Y.; received October 
8, 1937. 

** New York, N.Y. 


504 B. SCHLANGER [J. S. M. P. E. 

The artificiality of the present screen surroundings is further em- 
phasized in the presentation of color pictures, by the sharp contrast 
of the gray and black surroundings against the strong colors of the 
screen. In the case of black-and-white pictures, at least the black 
of the surroundings has some relation to the blacks or grays on the 
screen, although the intensities of the light on the surroundings and the 
screen vary to a disturbing degree. 

FIG. 1. Present form of screen picture, with jet black border surrounding 

the image. 

These present unnatural screen surroundings have had the effect 
of establishing a kind of cinematography that is most inflexible and 
limited in scope. For example, the black border around the screen 
has a tendency to force the use of pictures having low lighting inten- 
sities at the marginal areas, to avoid glaring contrasts at the border 
lines. Another example of the limits imposed is the hesitancy to place 
images of objects or human features near the sharp cut-off borders 
because of the unnatural effect of splitting the images. This hesi- 
tancy to use the marginal areas of the screen for various light inten- 


sities and image placements has resulted in reducing the effective 
area of the screen for action portrayal. This is most unfortunate 
when it is realized that the working screen area is not any too large 
for other than close-up shots. 

There are two suggested means by which this artificial limitation 
may be eliminated and a more lifelike projected picture be achieved. 
One is to project an enlarged picture upon a screen occupying a sub- 

FIG. 2. Screen synchrofield, showing luminous vignetting around the image. 

stantial portion of the spectator's field of vision. The other is to 
create an area contiguously surrounding the present screen, upon 
which a lighting effect can be imposed matching as nearly as possible 
the lighting occurring in the marginal areas of the picture. The 
former method apparently can not be ultilized at this time, due to 
limitations of the 35-mm. film and the present projection lighting 
and optical system, besides the difficulties encountered in the costs 
of producing larger settings to fill the larger screen, of which portions 
would necessarily have to be subdued to the main focal interest. 

506 B. SCHLANGER [J. S. M. P. E. 

The second means suggested would not require any change in the 
film width or in the screen size. The desired effect can be attained 
at and around the screen. Many attempts have been made in the 
past to create an illuminated field in line with or forward of the 
screen. These have all been unsuccessful because the source of il- 
lumination used was secondary, utilizing indirect lighting troughs 
around the picture. Although some attempts were made to vary 
the color and intensity of the light, it was practically impossible to 
create automatically colors and light intensities that would match 
the ever-changing colors and intensities occurring in the marginal 
areas of the picture. It is evident that a screen-border illumination 
of a fixed intensity and color can prove to be just as artificial and 
frame-creating as the present black border when the marginal areas 
of the picture are dark compared to the contiguous illuminated border. 

An illuminated field contiguous to the screen proper must not only 
have a constantly changing intensity of light and color, but its light 
and color must also vary along the four sides of the screen to match 
and blend the various edge conditions of the picture into the sur- 
rounding field. Fig. 1 shows a picture with the conventional black 
border. Fig. 2 shows the same picture with a contiguously surrounding 
field, having on it the various intensities nearly corresponding to the 
intensities existing on the picture margins. The various marginal 
light intensities have to blend into the background field illumination 
to render the physical edge of the screen as indefinite as possible. 

To meet the problems herein discussed, a system* devised by J. 
Gilston and the author will be demonstrated for the first time in 
conjunction with the presentation of this paper. The system util- 
izes the principle of employing the projection light-beam to create 
an illuminated field contiguous to the picture edges. It achieves the 
desired result of synchronizing automatically and with great sim- 
plicity, the color and intensity of light of the marginal areas of the 
picture with the color and light intensity of the surrounding field. 
Fig. 3 is a diagram of the screen and the arrangement of diffusing 
and reflecting surfaces forming a "screen synchrofield." The light 
falling from the projector upon the marginal areas of the screen is 
transmitted through the marginal areas of the screen upon the dif- 
fusing and reflecting surfaces behind and beyond the screen edges. 
Since the lighting of the screen marginal areas and the lighting 

* U. S. Pat. 

May, 1938] 



of the surrounding field have but one source, the necessary blending 
of the picture edge into the surrounding field is assured. No at- 
tempt is made to create upon the extended field any definition or 
duplication of forms occurring on the screen marginal areas. The 
extended field appears to have vague extensions of the color and 


'LIGHT im 




FIG. 3. 

Diagram of the screen 

light intensities of the screen marginal areas, thus simulating the 
effect of peripheral vision in real life. 

Elimination of the limiting artificial screen surroundings would 
help the spectator to connect himself more intimately with the space 
enfolded by the screen. The side walls of the motion picture theater 
auditorium can now be made to blend into the "screen synchrofield" 
surfaces, thereby making it further possible to "project" the spec- 
tator into the scene of action. For successful results, the walls must 
be designed to reject or receive light reflections from the screen to a 
proper degree. Secondary lighting must be completely eliminated 

508 B. SCHLANGER [J. S. M. P. E. 

from any of the auditorium surfaces at or near the screen, allowing 
only the carefully studied use of the screen lighting to control the 
lighting of the forepart of the theater. 

Another advantage of this procedure is the elimination of eye 
fatigue caused by the necessity of adjusting the eyes to accommodate 
the sharp contrast of the black border and the illuminated screen. 
Indirectly eye fatigue will also be reduced further, inasmuch as 
brighter levels of screen illumination may be used for improving 
visual acuity without the disadvantage of the glare created by high 
brightness levels within dark surroundings. 

The new and greater possibilities in cinematographic expression 
and new screen brightness evaluations resulting from the use of the 
' 'synchrofield' ' seem most encouraging. Eliminating the definite edge 
of the picture will, in effect, increase the effective area of the screen, 
since a freer use of the marginal areas of the screen for image place- 
ment will be possible. Panoramic views will be greatly enhanced by 
the apparent extension of the sky and nature's forms. The use of 
colored films will also be greatly enhanced by the elimination of the 
black surroundings that lend artificial hardness to the picture. Bril- 
liant colors on the picture proper will appear softer when brilliant 
colors in subdued degree appear in the peripheral screen surroundings. 
Achievement of actual depth effects in the projected motion picture 
will further demand the elimination of a sharp picture cut-off. The 
more realistically life is duplicated on the screen, the less it can afford 
to be cut off by abrupt and dark surroundings. 


MR. KELLOGG: How does the reflectivity of your translucent screen compare 
with that of a piece of good white paper? 

MR. SCHLANGER: There is no doubt that we are losing some of the light through 
the translucency. 

MR. KELLOGG: That is an indication of the price you pay in screen bright- 

MR. SCHLANGER: Where the incident light is the minimum possible, it would 
be necessary to increase the light upon the screen to compensate for the light 
transmitted through the screen. -The screen material used in this model is not 
considered the most desirable, and further experimenting is necessary to determine 
the most efficient material. 

MR. GOLDEN: Of what material is the field surrounding the screen? 

MR. SCHLANGER: This happens to be a white diffusive paper. Ordinary plaster 
would be more suitable. 


MR. KELLOGG: I was wondering whether you could not utilize some of the 
light reflected from the screen at angles too great to be useful for viewing the 
screen, say, inside of 45 degrees. I do not know whether there would be enough 
light reflected from the edges of the screen but it might be utilized, if sufficient, 
with less loss to the screen illumination, than to depend upon transmitted light. 

MR. SCHLANGER: I have investigated the possibility very thoroughly, and 
dropped the idea because if we allow the light entering the central area to reflect 
to a surface at, say, 45 degrees, there will still be sharp contrast between the pic- 
ture and the border. The purpose of this scheme is to blend only the edge con- 
dition into the surrounding field. 

MR. RICHARDSON: The scheme has been rejected before. 

MR. SCHLANGER: The conception embodied in this scheme differs fundamen- 
tally from previous proposed solutions. Earlier attempts employed fixed border 
illumination, while this arrangement rests upon synchronizing the field lighting 
with that of the screen edge. 

MR. RICHARDSON : I feel that when we put light outside the screen we detract 
from the picture. 

MR. SCHLANGER: That is true when the source of light near the picture and 
within the field of vision of the spectator is unrelated to the picture. But it is 
not objectionable if the light that appears within the field of vision is gauged to 
the edge light of the picture. The field illumination in this scheme operates as 
an integral part of the scene being viewed, and therefore does not detract from 
the picture. 

MR. GREENE: The light that passes through the edge of the screen and is re- 
flected to the outer edges of the border should not be regarded as light outside the 
screen. Psychologically, the plaster surface becomes part of the screen. How- 
ever, I wonder whether the area of the screen backed up by the theater speakers 
will cause non-uniformity of illumination of the picture area; in other words, the 
dark central portion where the speakers are located might appear much darker 
than the border of the active screen represented by the reflective surface. 

MR. MALMUTH: We manufacture a screen perforated in the center and solid or 
opaque at the border, and in tests we have made on a couple of hundred screens, 
we have found no case in which the masked area behind the loud speakers has 
been noticeable through the screen. 

MR. SCHLANGER: I do not think that answers Mr. Greene fully. The external 
marginal areas of the screen will not appear sufficiently brighter than the internal 
area, for two reasons: First, we know that the central area of the screen has a 
higher illumination level due to the optical system, and second, the reflectors are 
so designed as to reflect the light outwardly, rather than back to the screen again. 
This demonstration did not show a marked contrast between the marginal area 
that was being used and the central area. 

MR. CRABTREE: As one who has urged Mr. Schlanger in the past to proceed 
along these lines, I wish to congratulate him on having done something about 
which many of us have merely been thinking. It seems to me that the effective- 
ness of this scheme will be a maximum at a certain critical distance from the 
screen. I always like to sit close to the screen, where the margins are least con- 
spicuous. Farther back in the theater, the margins become very objectionable, 
and the effectiveness of this arrangement, I should think, would be a minimum 


at the back of the theater. Have you tried the effectiveness at various distances? 

MR. SCHLANGBR: Certainly for the major portion of the theater, where the 
range of vision takes in only the screen plus the synchrofield, the condition should 
be best. Farther back the angle of vision includes not only the screen and the 
synchrofield but some of the walls of the auditorium as well. To complete the 
idea, it would be necessary to continue the effective screen area still farther, and 
as stated in the paper, the auditorium walls and ceiling surfaces within the field 
of vision could be treated with a suitable quantity of light. The objection would 
then be overcome and the system would be effective for the complete depth of the 

MR. CRABTREB : Would not this kaleidoscopic movement occurring around the 
screen be a little distracting? It would depend, of course, upon the general il- 
lumination level in the theater. 

MR. SCHLANGBR: If this scheme were shown on a full-size screen the syn- 
chronized light field would, we believe, occupy the peripheral vision of the spec- 
tator, and therefore the synchronized light movements would hold a more natural 
and undisturbing effect, similar to the vague feeling of peripheral movement felt 
in real life. In the past few years I have built quite a few theaters and have 
dared to leave the wall surfaces quite bright, some of them even of white plaster. 
There has been a feeling that auditoriums ought to be pitch black, but I have 
found that if the walls and ceilings are evenly illuminated (that is very important), 
with no interruptions of dark or light areas, an astonishingly high illumination 
level can be put upon the walls and ceiling without detracting from the picture. 
On the other hand, in a darkroom with a picture being projected, as little as a 2- 
watt bulb behind a shield throwing a slight glimmer of light upon a dark area is 
very objectionable. What is required is an even bath of illumination over the 
complete surface, of an intensity that will blend from the screen lighting to your 
position in the auditorium. Objections to light in an auditorium are due to un- 
evenness of the light. 

MR. CRABTRBE: This is a matter for argument, and should be investigated. 
I hope you are successful hi persuading someone to install this idea in a theater 
and I am sure it will attract a great deal of interest. 

The screen is now suspended on a wire; why not cover the entire front of the 
proscenium with a transparent material and attach the screen in the middle? 
That would eliminate the black border. 

MR. SCHLANGER: Several methods of more or less completely eliminating the 
small hair-line dark edge visible around the screen are being considered. 



H. F. OLSON** 

Summary. The expressions for the throat impedance of a horn with two rates of 
exponential flare have been derived. This expression is applicable to any number of 
sections by considering two sections at a time. The impedance-frequency character- 
istic of specific multiple horns shows the possibility of obtaining a large variety of 
impedance characteristics suitable for improving the efficiency characteristic over that 
possible with a single rate of flare. The efficiency of a horn type of loud speaker hav- 
ing a horn with three rates of flare shows an efficiency within a few per cent of the 
ultimate efficiency. 

The efficiency of a horn loud speaker is governed, among many 
other factors, by the throat resistance. To obtain the maximum 
efficiency at any frequency, the effective reactance of the entire vi- 
brating system should be equal to the effective resistance. This, in 
general, means that to obtain maximum efficiency the throat resis- 
tance of the horn should be proportional to the frequency, since the 
reactance is primarily mass reactance and, therefore, proportional to 
the frequency. It is well known that the surge resistance of an ex- 
ponential horn is independent of the frequency over its transmission 
range. Because of this fact, in order to obtain high efficiency over 
a wide range, horn loud speakers have been built using two or more 
units to cover the range of reproduction; that is, a relatively large 
throat horn for the reproduction of low frequencies and a relatively 
small throat horn for the reproduction of high frequencies. Obvi- 
ously, the same result could be attained with a horn in which the 
throat resistance increased at the proper rate with respect to the 
frequency. Such a system would eliminate the use of two or more 
mechanisms, cross-over networks, phase-shift between units due to 
the filter, and different sound-path lengths, and other problems 
connected with multiple channel systems. It is the purpose of this 
paper to outline a method for obtaining practically any throat im- 

* Received January 18, 1938. 
** RCA Manufacturing Co., Camden, N. J. 




[J. S. M. P. E. 




m m 












U. 4 
U) 3 


V s 


v \ 



t * 

lO 1 * 

fc 7 

'I0 3 

4 . 

\0" l 


FIG. 1. Efficiency characteristics of a diaphragm 
coupled to a pure resistance and driven by an aluminum 
coil of one-half the diaphragm mass operating in a field of 
22,000 gauss. 

pedance-frequency characteristic by employing a horn consisting of 
manifold exponential sections. 

The efficiency of a simple system consisting of a dynamically driven 
diaphragm coupled to an acoustic resistance is given by 

Eff = 


where r\i = real part of the mechanical impedance Z M - 

ZM = 

B = 

/ = 

Z T = 

RJU = 

m = 

A = 

r D = 


flux density. 

length of voice-coil conductor. 

AR\i + jum. 

acoustic resistance. 

mass of diaphragm and coil. 

area of diaphragm. 

damped resistance of voice-coil. 











s. 20 










FIG. 2. Efficiency characteristic of a dynamically 
driven diaphragm coupled to an acoustic resistance: 
A, without air chamber; B, with air-chamber. 

May, 1938] 



The efficiency characteristics of this system for a voice-coil of half the 
mass of the diaphragm operating in a gap of 22,000 gauss for various 
initial efficiencies is shown in Fig. 1 . These characteristics show that 
to attain a high efficiency at the high frequencies a relatively large 
throat resistance is required, while to attain a high efficiency at the 
low frequencies a relatively small throat resistance is required. 

The discussion has assumed that the system consists of only two 
elements, namely, the throat resistance and diaphragm plus coil 
mass. The stiffness of the suspension system influences the low- 



o * 


O 1 

I0 3 



FIG. 3. Throat impedance characteristic of an ex- 
ponential horn of the dimensions shown and cut off due 
to flare of 40 cycles: R, resistive component; X, re- 
active component. 

frequency efficiency. In general, the stiffness is chosen so that the 
capacitance annuls the large positive reactance near the cut-off of a 
finite horn. The capacitance of the air-chamber affects the effi- 
ciency at the high frequencies. As a matter of fact, the air-chamber 
is extremely useful for increasing the efficiency and effecting a sharp 

The efficiency characteristics of a system consisting of a dynami- 
cally driven cone coupled to a resistance with and without an air- 
chamber are shown in Fig. 2. These characteristics show that it is 
possible to improve the shape of the efficiency characteristic by means 
of an air chamber. 



[J. S. M. P. E. 

To proceed with the subject of a method for obtaining a horn 
system that will make it possible to attain practically the maximum 
efficiency throughout the range, let us consider first a finite exponen- 
tial horn. The throat acoustic impedance 1 is given by 

- 00 + ^ sin (bill) 








O 4 

FIG. 4. Throat impedance characteristic of a multiple 
flare exponential horn. The large horn is the same as 
shown in Fig. 3. The small horn has a flare cut-off of 
600 cycles: R, resistive component at throat of small 
horn; X, reactive component. In this example the 
impedance is referred to 3 while in Fig. 3 the impedance 
is referred to 5 2 . 

density of air. 
velocity of sound, 
flare constant. 

bi = % \/ 
K = 2T/X. 

May, 1938] 



X = wavelength. 

w = 27rf. 

/ = frequency. 

li = length of the horn. 

0i = tan" 1 

Sz = throat area. 
S-2 = mouth area. 
Zi = mouth acoustic impedance. 

A typical impedance characteristic for a horn of this type is shown 
in Fig. 3. 

Suppose that a short horn having a high rate of flare compared to 
that of the large horn is connected to the large horn of Fig. 3, as shown 
in Fig. 4. The impedance at the throat of the small horn is given by 

Z 2 cos (b 2 h - 00 + * sin ( W 

jZ 2 sin (62/2) + cos (62/2 + 

where a 2 

m> t 

= the flare constant of small horn. 
= Y 2 V 4K* - wV 
= length of small horn. 

02 = tan" 1 ' 


S a = throat area of small horn. 
5 2 = mouth area of large horn. 
Z 2 = acoustic impedance obtained from equation 1. 

For b z = equation 3 is indeterminate. To evaluate, take the 
derivative of the numerator and denominator with respect to b z> and 
set 6 2 = 0. Then equation 3 becomes: 



In most cases for the frequency at which & = 0, the impedance at the 
mouth of the small horn is a constant resistance of the value. 



[J. S. M. P. E. 

Then the expression for the impedance at the throat of the small horn 
becomes, for b z = : 







a? .s 


I0 1 jo 1 


FIG. 5. Throat impedance characteristic of a multiple 
flare exponential horn of three sections. The cut-offs 
due to flare of the three horns are 25, 100, and 1400 
cycles: R, resistive component at the throat of the 
small horn; X, reactive component. In this example 
the impedance is referred to 6Y 

Below the frequency corresponding to bi = 0, bi is imaginary. This 
portion of the range may be evaluated by employing the standard 
formulas involving complex quantities. 

The impedance characteristic of a horn with two rates of flare is 
shown in Fig. 4. The large horn is the same as that shown in Fig. 3. 
This characteristic shows that at the low frequencies the impedance 

May, 1938] 



characteristic is the same as that shown in Fig. 3. However, at the 
high frequencies the resistance is more than six times the resistance 
at the low frequencies. Employing this horn it is possible to attain 
practically the same efficiency characteristic as in the case of two 
separate horns and driving mechanisms. 

Fig. 5 shows the impedance characteristic of a horn consisting of 
three exponential sections. The three distinct steps representing the 
three horns are quite evident. 

The efficiency characteristic of a diaphragm and coil, having a mass 
ratio of 2, operating in a field of 22,000 gauss, coupled to the horn of 
Fig. 5, is shown in Fig. 6. This efficiency characteristic is only a 












* , 



k *. 

u. fc 








FIG. 6. Efficiency characteristic of a diaphragm 
coupled to the horn of Fig. 5 and driven by an aluminum 
coil of one-half the diaphragm mass operating in a field 
of 22,000 gauss: A, without air chamber; B, with air- 

few per cent below the ultimate efficiency characteristic obtained 
from the envelope of the family of characteristics shown in Fig. 1. 
The characteristic A shown in Fig. 6 was computed assuming the 
capacitance of the air chamber to be zero. Of course such a condition 
is impossible in a practical loud speaker. The dotted characteristic B 
of Fig. 6 shows the efficiency with an air-chamber. The air-chamber 
increases the efficiency over a considerable range and effects a sharp 
cut-off at the high-frequency response limit. 

High-power horn loud speakers for sound reenforcing and public 
address systems employing multiple flare horns have been developed 
by M. L. Graham. 2 The high efficiency over a wide frequency 
range exhibited by these loud speakers confirms the theoretical 
analysis outlined above. 

518 H. F. OLSON 

The above examples have shown how it is possible to obtain a 
large variety of impedance characteristics by employing multiple 
exponential horns with different rates of flare. To attain maximum 
efficiency at any frequency, the effective reactance of the entire vi- 
brating system should be equal to the effective radiation resistance. 
The principal reactance* at the high frequencies is due to the mass of 
the diaphragm. If the throat resistance increases with frequency as 
in the case of multiple exponential horns, it is possible to improve the 
efficiency as shown in the examples above. Of course, it is not neces- 
sary to employ exponential horns. The same results could be at- 
tained with a horn having a predetermined rate of expansion to yield 
the desifed resistance characteristic. Multiple exponential horns 
have been used in this paper to illustrate the principles involved. 


1 OLSON, H. F., AND MASSA, F. : "Applied Acoustics," P. Blakiston's Son Co. 
(Philadelphia), p. 188. 

1 GRAHAM, M. L.: "New High-Powered Sound Projectors," Broadcast News 
(Dec., 1937). No. 27, p. 4. 

* The horn throat impedance has a reactive as well as a resistive component. 
In general the constants of the system can be chosen so that this reactance will not 
materially reduce the efficiency. 



Summary. Design requirements for the construction of scoring stages are dis- 
cussed, and, after a brief examination of the uses to which such recording stages are 
put in motion picture studios, there follows an examination of their most desirable 
shape and of the amount of sound-insulation necessary for their walls. The matter 
of optimal reverberation is investigated from the standpoints of the variation of rever- 
beration with frequency, of accommodating different musical performances by pro- 
viding control of the reverberation, and of considering the ratio of "initial" sound to 
generally reflected sound an extension of the term "recorded reverberation." 

The adaptation of musical accompaniment to a motion picture may 
be effected in a variety of ways. Sound-tracks of vocalists accom- 
panied by an orchestra may be made, and a print of the recording 
played back when the scene is photographed. Because the cameras 
are electrically interlocked with the sound reproducers, and picture 
cuts will be printed with the original sound-track, action and sound 
on the screen are in complete synchronism. This method is known 
as pre-scoring. Again, a picture may be shown upon the screen of 
a stage in which an orchestra is assembled. During the showing of 
a sequence, which might carry only a piano track, a recording of the 
orchestra is made while the musical director listens to the sound- 
track through a pair of ear-phones. Later the piano track and the 
orchestra track are combined by a method of re-recording, or dubbing, 
and then cut into a finished picture. Taking the original sequence 
with its piano track is known as synchronous or direct recording with 
incomplete accompaniment; making the orchestra track is referred 
to as post-scoring (sometimes also as merely scoring). There are, 
of course, other ways in which music may be set to a picture, but the 
above illustration should indicate the complexities of the task. 

The studio in which the music is recorded is known as "recording" 
or "scoring stage," in contradistinction to a "sound-stage," which is 
used for recording action scenes. Only rarely is photography done 

* Received December 28, 1937. 
** RCA Manufacturing Company, Hollywood, Calif. 




[J. S. M. P. E. 

in a scoring stage. If dialog is recorded there, it is mostly in the 
form of commentation to newsreels, travelogues, and trailers. Scor- 
ing stages thus differ from the smaller radio studios used for a variety 
of broadcasts, although the acoustic principles underlying the design 
of the latter can with some modifications successfully be carried over 
to the design of the former. 

Design Considerations. The problems associated with the de- 
sign of a recording stage are manifold. Since musical performances 
rendered in a scoring stage vary so widely as to type of program and 


Wire Netting Fiberboard 

Building paper/ Ai r .space 
-I" Wood Sheathing / Sheetrook 

FIG. 1. Section through well insulating wall 
for a recording studio. 

number of artists involved, such rooms should permit some control 
of the acoustics. The large size of the stage, necessary to accommo- 
date symphony orchestras, requires careful consideration of reverbera- 
tion, echoes, and interference. Again, the wider range of the audio- 
frequency spectrum recorded in such a stage necessitates provision 
for sustaining and distributing the higher notes as much as possible. 
Several ratios for the dimensions of height, width, and length of 
recording stages have been proposed. Many of the larger American 
broadcast studios have a ratio of 2:3:5^ while in Germany the ratio 
is 1:2:3. The British Broadcasting Company 2 advocates that the 
length be made 25 to 75 per cent greater than the breadth, and that 
considerable height is a definite advantage. Of the ratios 2:3:5 

May, 1938] 



and 1:2:3, for studios of equal volume, the writer favors the latter, first 
because it makes for greater ceiling height; and second, because the 
mean free path* is somewhat shorter. As is well known, the ratio 
of velocity of sound to mean free path represents the average number 
of reflections per second at any point in the room, so that a shorter 
mean free path will make for a larger number of such reflections per 
second, and hence for a more diffused state of sound in the room. 

Rubber Pads 
Wood Support 
Acoustlo Material 

FIG. 2. 

Section through monitoring-room 

Patently, for very large stages a compromise must be made as to 
ceiling height, to prevent its becoming abnormally high. 

Next in order of design considerations may be discussed the prob- 
lem of adequate insulation against outside noise, both air-borne and 
solid-borne. The subject is extensive, and many papers and books 
have been written about it. Here only special remarks pertaining 
to sound stages will be made, with particular attention to the present 
trend toward non-concrete structures. 

* The mean free path for a rectangular room is given by 4 V/s, where V is the 
volume and 5 the total interior surface of the room. 



[J. S. M. P. E. 

After a noise survey has been conducted at the proposed site, a 
type of construction should be selected that will result in a noise-level 
within the stage of not more than 30 db. above the reference stand- 
ard of 0.0002 dyne/cm. 2 as measured with a noise-level meter 3 having 
a 30-db. equi-loudness contour characteristic. An attempt, on the 
other hand, to establish an extraordinarily low noise level in the stage 
leads either to massive concrete or to very thick and elaborately con- 
structed multi-layer walls, which will prove very expensive. It may 
be kept in mind that the thermal agitation noise of microphones is 




I 1<0 


8 .8